Files
frontier-gpu-bandwidth/results/implicit_managed_GPURdHost_fine.json
2023-10-09 17:04:07 -04:00

25273 lines
893 KiB
JSON

{
"context": {
"date": "2023-10-09T14:27:09-04:00",
"host_name": "frontier10485",
"executable": "/lustre/orion/csc465/scratch/cpearson/frontier-gpu-bandwidth/build/comm_scope",
"num_cpus": 128,
"mhz_per_cpu": 1728,
"cpu_scaling_enabled": false,
"caches": [
{
"type": "Data",
"level": 1,
"size": 32768,
"num_sharing": 2
},
{
"type": "Instruction",
"level": 1,
"size": 32768,
"num_sharing": 2
},
{
"type": "Unified",
"level": 2,
"size": 524288,
"num_sharing": 2
},
{
"type": "Unified",
"level": 3,
"size": 33554432,
"num_sharing": 16
}
],
"load_avg": [2.12,2.56,2.44],
"library_build_type": "release"
},
"benchmarks": [
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"family_index": 0,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 378,
"real_time": 1.8628192977173186e+06,
"cpu_time": 2.0664527486772484e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.1988176765289041e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"family_index": 0,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 378,
"real_time": 1.8669374079012962e+06,
"cpu_time": 2.0662510767195774e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.1939675013553281e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"family_index": 0,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 378,
"real_time": 1.8616810934126892e+06,
"cpu_time": 2.0656471375661376e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2001620011575297e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"family_index": 0,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 378,
"real_time": 1.8662707093065523e+06,
"cpu_time": 2.0651398915343892e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.1947512649555248e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"family_index": 0,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 378,
"real_time": 1.8700773003084583e+06,
"cpu_time": 2.0694383518518514e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.1902837916509593e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8655571617292627e+06,
"cpu_time": 2.0665858412698410e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.1955964471296491e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time_median",
"family_index": 0,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8662707093065523e+06,
"cpu_time": 2.0662510767195772e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.1947512649555248e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.3677027538529051e+03,
"cpu_time": 1.6761960537402726e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.9627234291984546e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.8051994454735658e-03,
"cpu_time": 8.1109432778766739e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.8048505381665237e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"family_index": 0,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 378,
"real_time": 1.8654093254693579e+06,
"cpu_time": 2.0657390846560840e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3915294558414416e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"family_index": 0,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 378,
"real_time": 1.8707871068924698e+06,
"cpu_time": 2.0711520185185184e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3789055258177295e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"family_index": 0,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 378,
"real_time": 1.8606127327575691e+06,
"cpu_time": 2.0655969894179872e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4028506608459223e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"family_index": 0,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 378,
"real_time": 1.8650939872179786e+06,
"cpu_time": 2.0656543465608438e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3922719477635520e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"family_index": 0,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 378,
"real_time": 1.8695003439716331e+06,
"cpu_time": 2.0702673703703738e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3819194933105083e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8662806992618020e+06,
"cpu_time": 2.0676819619047616e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3894954167158306e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time_median",
"family_index": 0,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8654093254693579e+06,
"cpu_time": 2.0657390846560840e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3915294558414416e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.0301130208669406e+03,
"cpu_time": 2.7820287602332965e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 9.4835303328055106e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.1594356210515556e-03,
"cpu_time": 1.3454819510397402e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.1605058059044465e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"family_index": 0,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 377,
"real_time": 1.8673297063938824e+06,
"cpu_time": 2.0700095968169754e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7740263242746610e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"family_index": 0,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 377,
"real_time": 1.8710228408389138e+06,
"cpu_time": 2.0744616366047764e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7567076373337470e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"family_index": 0,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 377,
"real_time": 1.8674654849302839e+06,
"cpu_time": 2.0733772705570280e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7733883877439629e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"family_index": 0,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 377,
"real_time": 1.8719552528772091e+06,
"cpu_time": 2.0777826737400547e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7523459627668280e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"family_index": 0,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 377,
"real_time": 1.8714756744006521e+06,
"cpu_time": 2.0778288859416440e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7545888114453033e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8698497918881886e+06,
"cpu_time": 2.0746920127320960e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7622114247129019e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time_median",
"family_index": 0,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8710228408389140e+06,
"cpu_time": 2.0744616366047766e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7567076373337470e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.2631976452603608e+03,
"cpu_time": 3.2824837374181725e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.0609425080883271e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.2103633431298066e-03,
"cpu_time": 1.5821547088792107e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2108159192507609e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"family_index": 0,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 377,
"real_time": 1.8702584853570596e+06,
"cpu_time": 2.0780134217506647e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7520572828062378e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"family_index": 0,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 377,
"real_time": 1.8746442687723956e+06,
"cpu_time": 2.0829402015915138e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7479582951200664e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"family_index": 0,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 377,
"real_time": 1.8701570627924339e+06,
"cpu_time": 2.0782183687002666e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7521523005705360e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"family_index": 0,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 377,
"real_time": 1.8741485756515916e+06,
"cpu_time": 2.0822133713527906e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7484206122029275e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"family_index": 0,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 377,
"real_time": 1.8749392845168270e+06,
"cpu_time": 2.0826716047745442e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7476832594312157e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8728295354180620e+06,
"cpu_time": 2.0808113936339561e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7496543500261970e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time_median",
"family_index": 0,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8741485756515916e+06,
"cpu_time": 2.0822133713527906e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7484206122029275e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.4102164787451470e+03,
"cpu_time": 2.4753876109039179e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.2526513409896572e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.2869385243901160e-03,
"cpu_time": 1.1896261326120809e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2874836340994603e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"family_index": 0,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 377,
"real_time": 1.8705628358463163e+06,
"cpu_time": 2.0840395862068955e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5035444275973193e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"family_index": 0,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 377,
"real_time": 1.8772365877505785e+06,
"cpu_time": 2.0905449469495974e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.4910889989912942e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"family_index": 0,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 377,
"real_time": 1.8709626312088778e+06,
"cpu_time": 2.0845369363395195e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5027957751168706e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"family_index": 0,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 377,
"real_time": 1.8779292090452246e+06,
"cpu_time": 2.0921545225464161e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.4898014091446906e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"family_index": 0,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 377,
"real_time": 1.8702097269070314e+06,
"cpu_time": 2.0842613262599451e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5042059217809759e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8733801981516059e+06,
"cpu_time": 2.0871074636604753e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.4982873065262303e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time_median",
"family_index": 0,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8709626312088780e+06,
"cpu_time": 2.0845369363395195e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5027957751168706e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.8535467118190159e+03,
"cpu_time": 3.9181971503318514e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.1906137298912116e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.0570019452651233e-03,
"cpu_time": 1.8773336871978402e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.0554668898911652e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"family_index": 0,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 380,
"real_time": 1.8728798400077624e+06,
"cpu_time": 2.0981805763157867e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 6.9984201442126021e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"family_index": 0,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 380,
"real_time": 1.8670808081591723e+06,
"cpu_time": 2.0974538973684269e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0201567831029758e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"family_index": 0,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 380,
"real_time": 1.8757696530404869e+06,
"cpu_time": 2.1018046947368416e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 6.9876383695376337e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"family_index": 0,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 380,
"real_time": 1.8673089928358279e+06,
"cpu_time": 2.0971973078947351e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0192989217571735e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"family_index": 0,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 380,
"real_time": 1.8720268951591081e+06,
"cpu_time": 2.0973531552631585e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0016088090903133e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8710132378404713e+06,
"cpu_time": 2.0983979263157896e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0054246055401400e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time_median",
"family_index": 0,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8720268951591079e+06,
"cpu_time": 2.0974538973684269e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0016088090903133e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.7523534153405540e+03,
"cpu_time": 1.9414900139252593e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4049080738610623e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.0055194369824614e-03,
"cpu_time": 9.2522490113873784e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.0054574175989428e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"family_index": 0,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 379,
"real_time": 1.8719649071744359e+06,
"cpu_time": 2.1206745356200528e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4003681318774450e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"family_index": 0,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 379,
"real_time": 1.8759729443726162e+06,
"cpu_time": 2.1203973430079189e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.3973762296858132e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"family_index": 0,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 379,
"real_time": 1.8796512702652584e+06,
"cpu_time": 2.1242366306068548e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.3946416771394300e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"family_index": 0,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 379,
"real_time": 1.8788765876162010e+06,
"cpu_time": 2.1238162348284936e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.3952167041082332e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"family_index": 0,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 379,
"real_time": 1.8791353774175914e+06,
"cpu_time": 2.1242754617414167e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.3950245583703095e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8771202173692207e+06,
"cpu_time": 2.1226800411609476e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.3965254602362463e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time_median",
"family_index": 0,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8788765876162010e+06,
"cpu_time": 2.1238162348284936e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.3952167041082332e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.2189048511765131e+03,
"cpu_time": 1.9679962467844557e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.3981306139366137e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.7148101764562530e-03,
"cpu_time": 9.2712806858451864e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.7172122401054749e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"family_index": 0,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 369,
"real_time": 1.9024523606222265e+06,
"cpu_time": 2.1845281815718277e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7558535017850512e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"family_index": 0,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 369,
"real_time": 1.9027803509324528e+06,
"cpu_time": 2.1854544471544656e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7553784636417651e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"family_index": 0,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 369,
"real_time": 1.9027395724435020e+06,
"cpu_time": 2.1846839485094836e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7554375154278642e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"family_index": 0,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 369,
"real_time": 1.9006010313419914e+06,
"cpu_time": 2.1838460325203254e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7585379117142045e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"family_index": 0,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 369,
"real_time": 1.8969101671679679e+06,
"cpu_time": 2.1813700325203235e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7639052659132874e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9010966965016283e+06,
"cpu_time": 2.1839765284552854e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7578225316964340e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time_median",
"family_index": 0,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9024523606222265e+06,
"cpu_time": 2.1845281815718277e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7558535017850512e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.5072031540156418e+03,
"cpu_time": 1.5652362615521006e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.6417809579929436e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.3188193733803032e-03,
"cpu_time": 7.1669097225105427e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.3205276685272259e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"family_index": 0,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 370,
"real_time": 1.9021051190950831e+06,
"cpu_time": 2.2657980675675552e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5127132011445022e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"family_index": 0,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 370,
"real_time": 1.8997820934305927e+06,
"cpu_time": 2.2639750513513484e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5194540659476388e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"family_index": 0,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 370,
"real_time": 1.9044956182935147e+06,
"cpu_time": 2.2663902837837827e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5057937121407270e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"family_index": 0,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 370,
"real_time": 1.8992069846717641e+06,
"cpu_time": 2.2632703486486604e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5211254405807865e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"family_index": 0,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 370,
"real_time": 1.9049124882676061e+06,
"cpu_time": 2.2664321459459425e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5045888273513901e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9021004607517119e+06,
"cpu_time": 2.2651731794594582e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5127350494330096e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time_median",
"family_index": 0,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9021051190950833e+06,
"cpu_time": 2.2657980675675557e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5127132011445022e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.6168363992322329e+03,
"cpu_time": 1.4588609216729265e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.5843164074292150e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.3757614033688087e-03,
"cpu_time": 6.4403946457685713e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.3757810486845852e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"family_index": 0,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 361,
"real_time": 1.9540153635688995e+06,
"cpu_time": 2.4688007617728710e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0732525644883721e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"family_index": 0,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 361,
"real_time": 1.9504590175395776e+06,
"cpu_time": 2.4647507562326831e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0752094666646571e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"family_index": 0,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 361,
"real_time": 1.9495757019348505e+06,
"cpu_time": 2.4647675761772953e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0756966235877314e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"family_index": 0,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 361,
"real_time": 1.9480754546189057e+06,
"cpu_time": 2.4640008698060936e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0765250365573020e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"family_index": 0,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 361,
"real_time": 1.9467963291773277e+06,
"cpu_time": 2.4577503878116473e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0772323578842010e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9497843733679124e+06,
"cpu_time": 2.4640140703601181e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0755832098364530e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time_median",
"family_index": 0,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9495757019348505e+06,
"cpu_time": 2.4647507562326831e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0756966235877314e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.7499444530560027e+03,
"cpu_time": 3.9767067119393605e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.5158480488492241e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.4103838817345498e-03,
"cpu_time": 1.6139139624954174e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4093266192577655e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"family_index": 0,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 342,
"real_time": 2.0597699787249865e+06,
"cpu_time": 2.8785452456140271e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0362972775223701e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"family_index": 0,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 342,
"real_time": 2.0599131242982086e+06,
"cpu_time": 2.8796132982456163e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0361557730396795e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"family_index": 0,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 342,
"real_time": 2.0572600111144744e+06,
"cpu_time": 2.8785397953216354e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0387816694729950e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"family_index": 0,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 342,
"real_time": 2.0580333967322800e+06,
"cpu_time": 2.8787064590643304e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0380155184360294e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"family_index": 0,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 342,
"real_time": 2.0556787480954190e+06,
"cpu_time": 2.8781602514619753e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0403499349720922e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0581310517930738e+06,
"cpu_time": 2.8787130099415169e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0379200346886334e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time_median",
"family_index": 0,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0580333967322796e+06,
"cpu_time": 2.8785452456140267e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0380155184360294e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.7779135031814235e+03,
"cpu_time": 5.4181455562995995e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.7608170703397896e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 8.6384853949532489e-04,
"cpu_time": 1.8821416159194257e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.6402657629734648e-04,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"family_index": 0,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 329,
"real_time": 2.1137411884487942e+06,
"cpu_time": 3.6210186352583473e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9686069637296162e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"family_index": 0,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 329,
"real_time": 2.1204835587353827e+06,
"cpu_time": 3.6252245258358582e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9559882298747039e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"family_index": 0,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 329,
"real_time": 2.1215714758409238e+06,
"cpu_time": 3.6237042249240056e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9539596452554216e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"family_index": 0,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 329,
"real_time": 2.1223651594128516e+06,
"cpu_time": 3.6237562978723450e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9524810152465434e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"family_index": 0,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 329,
"real_time": 2.1191559626514199e+06,
"cpu_time": 3.6242782826747820e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9584665535916681e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.1194634690178749e+06,
"cpu_time": 3.6235963933130680e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9579004815395908e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time_median",
"family_index": 0,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.1204835587353827e+06,
"cpu_time": 3.6237562978723450e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9559882298747039e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.4184476746944410e+03,
"cpu_time": 1.5650736660713665e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.3934871235281294e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.6128835078617772e-03,
"cpu_time": 4.3191169661155717e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.6153733913595310e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"family_index": 0,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 191,
"real_time": 3.6690382381988913e+06,
"cpu_time": 6.6208500994764548e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5726468111806412e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"family_index": 0,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 191,
"real_time": 3.6646076831576554e+06,
"cpu_time": 6.6229438586387513e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5781751965175438e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"family_index": 0,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 191,
"real_time": 3.6663928542149630e+06,
"cpu_time": 6.6235126858638870e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5759460775493708e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"family_index": 0,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 191,
"real_time": 3.6693583145074509e+06,
"cpu_time": 6.6257190209423909e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5722479414638624e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"family_index": 0,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 191,
"real_time": 3.6672347759174583e+06,
"cpu_time": 6.6208939528795788e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5748955344160433e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6673263731992841e+06,
"cpu_time": 6.6227839235602133e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5747823122254925e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time_median",
"family_index": 0,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6672347759174583e+06,
"cpu_time": 6.6229438586387504e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5748955344160433e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9577021321532980e+03,
"cpu_time": 2.0299713731159161e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.4424492164055323e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 5.3382271795063816e-04,
"cpu_time": 3.0651330264519082e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.3389408494441677e-04,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"family_index": 0,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 96,
"real_time": 7.4580427559946356e+06,
"cpu_time": 1.3661284249999905e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4990935420730295e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"family_index": 0,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 96,
"real_time": 7.4633144201167552e+06,
"cpu_time": 1.3685648604166694e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4959156362964916e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"family_index": 0,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 96,
"real_time": 7.4622077809181064e+06,
"cpu_time": 1.3696042906249890e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4965823768407125e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"family_index": 0,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 96,
"real_time": 7.4593094420076041e+06,
"cpu_time": 1.3718653770833366e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4983295385275154e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"family_index": 0,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 96,
"real_time": 7.4559328592537595e+06,
"cpu_time": 1.3697755499999965e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5003667057375240e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.4597614516581725e+06,
"cpu_time": 1.3691877006249964e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4980575598950548e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time_median",
"family_index": 0,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.4593094420076041e+06,
"cpu_time": 1.3696042906249892e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4983295385275154e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.0176339658508596e+03,
"cpu_time": 2.0878315667751136e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.8195732823714465e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 4.0452150989092194e-04,
"cpu_time": 1.5248687713321383e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.0452423254759313e-04,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"family_index": 0,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 48,
"real_time": 1.4645315493301800e+07,
"cpu_time": 2.6583978145833537e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5822750647258501e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"family_index": 0,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 48,
"real_time": 1.4358375493126610e+07,
"cpu_time": 2.5624060833333280e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6738479594801779e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"family_index": 0,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 48,
"real_time": 1.4660928922239691e+07,
"cpu_time": 2.6587644395833421e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5773950856688318e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"family_index": 0,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 48,
"real_time": 1.4629838774756839e+07,
"cpu_time": 2.6576874208333358e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5871225946654634e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"family_index": 0,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 48,
"real_time": 1.4638855548885962e+07,
"cpu_time": 2.6584311687500119e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5842971655736494e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4586662846462181e+07,
"cpu_time": 2.6391373854166750e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6009875740227947e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time_median",
"family_index": 0,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4638855548885962e+07,
"cpu_time": 2.6583978145833544e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5842971655736494e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.2812000005459548e+05,
"cpu_time": 4.2895897226946254e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.0884614058075197e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 8.7833661066396302e-03,
"cpu_time": 1.6253756800983562e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.8860518313307324e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"family_index": 0,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 25,
"real_time": 2.8248011246323586e+07,
"cpu_time": 4.9986145400000058e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7514045087853088e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"family_index": 0,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 25,
"real_time": 2.8171223923563957e+07,
"cpu_time": 5.0085204080000378e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7643555837037287e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"family_index": 0,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 25,
"real_time": 2.8239275068044662e+07,
"cpu_time": 5.0077168599999599e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7528744160957479e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"family_index": 0,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 25,
"real_time": 2.8187588974833488e+07,
"cpu_time": 5.0072793200000092e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7615895108954725e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"family_index": 0,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 25,
"real_time": 2.7786782085895538e+07,
"cpu_time": 4.8496887599999927e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.8302724505882378e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.8126576259732254e+07,
"cpu_time": 4.9743639776000015e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7720992940136995e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time_median",
"family_index": 0,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.8187588974833488e+07,
"cpu_time": 5.0072793200000100e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7615895108954725e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9275860760632119e+05,
"cpu_time": 6.9811345653269440e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.2986386597295556e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 6.8532552923011233e-03,
"cpu_time": 1.4034225474379452e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.9123428841212330e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"family_index": 0,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 10,
"real_time": 5.4552702233195305e+07,
"cpu_time": 9.3672381099999309e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9206628638215666e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"family_index": 0,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 10,
"real_time": 5.4461663216352463e+07,
"cpu_time": 9.3779136599999905e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9288883252357330e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"family_index": 0,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 10,
"real_time": 5.0718251988291740e+07,
"cpu_time": 8.9949174599999532e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.2926795675444031e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"family_index": 0,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 10,
"real_time": 5.4467422887682915e+07,
"cpu_time": 9.3814714299999475e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9283671187002888e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"family_index": 0,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 10,
"real_time": 5.4440735653042793e+07,
"cpu_time": 9.3578731399999529e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9307830392074547e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.3728155195713043e+07,
"cpu_time": 9.2958827599999562e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.0002761829018898e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time_median",
"family_index": 0,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.4461663216352463e+07,
"cpu_time": 9.3672381099999309e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9288883252357330e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.6831307718034324e+06,
"cpu_time": 1.6849999553220223e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.6350417172232494e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 3.1326792548010823e-02,
"cpu_time": 1.8126303857580389e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.2699028161967639e-02,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"family_index": 0,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 8,
"real_time": 1.0655531473457813e+08,
"cpu_time": 1.8058430624999923e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0384245341239719e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"family_index": 0,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 8,
"real_time": 9.4823699444532394e+07,
"cpu_time": 1.7365919262500107e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.6617798624704084e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"family_index": 0,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 8,
"real_time": 1.0617835167795420e+08,
"cpu_time": 1.7996775199999958e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0563123604363737e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"family_index": 0,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 8,
"real_time": 9.4792198389768600e+07,
"cpu_time": 1.7452520012500018e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.6636613679164038e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"family_index": 0,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 8,
"real_time": 1.0616035200655460e+08,
"cpu_time": 1.8008009787500077e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0571696669473391e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.0170198325067759e+08,
"cpu_time": 1.7776330977500018e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.2954695583788996e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time_median",
"family_index": 0,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.0616035200655460e+08,
"cpu_time": 1.7996775199999958e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0571696669473391e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 6.2953488325637924e+06,
"cpu_time": 3.3732115788200079e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.3533699302803844e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 6.1899961351263522e-02,
"cpu_time": 1.8975859434039417e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.3325261212661008e-02,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"family_index": 0,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 4,
"real_time": 1.9139072671532631e+08,
"cpu_time": 3.0807746949999881e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.6102081977936096e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"family_index": 0,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 4,
"real_time": 1.9146832451224327e+08,
"cpu_time": 3.0902189925000203e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.6079345068449717e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"family_index": 0,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 4,
"real_time": 1.9113072752952576e+08,
"cpu_time": 3.0917194349999732e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.6178398830932560e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"family_index": 0,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 4,
"real_time": 1.9208712503314018e+08,
"cpu_time": 3.0960023975000441e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.5898687838383274e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"family_index": 0,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 4,
"real_time": 1.9281717017292976e+08,
"cpu_time": 3.0913404999999726e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.5687044003239193e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time_mean",
"family_index": 0,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9177881479263306e+08,
"cpu_time": 3.0900112039999998e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.5989111543788166e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time_median",
"family_index": 0,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9146832451224327e+08,
"cpu_time": 3.0913404999999726e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.6079345068449717e+09,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time_stddev",
"family_index": 0,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 6.7828290019170276e+05,
"cpu_time": 5.6110550505858310e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.9758247049954608e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time_cv",
"family_index": 0,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/0/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 3.5367978518644912e-03,
"cpu_time": 1.8158688367609659e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.5289445581757456e-03,
"dst_gpu": NaN,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"family_index": 1,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 379,
"real_time": 1.8632594538784383e+06,
"cpu_time": 2.0649631609497985e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.1982982517405380e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"family_index": 1,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 379,
"real_time": 1.8607154964284459e+06,
"cpu_time": 2.0623770131926078e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2013037500155587e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"family_index": 1,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 379,
"real_time": 1.8656902797288252e+06,
"cpu_time": 2.0673825013193106e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.1954340677571343e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"family_index": 1,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 379,
"real_time": 1.8611262661445148e+06,
"cpu_time": 2.0625407546174033e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2008178996287133e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"family_index": 1,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 379,
"real_time": 1.8565470491875992e+06,
"cpu_time": 2.0623606253298142e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2062462687343992e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8614677090735647e+06,
"cpu_time": 2.0639248110817871e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2004200475752684e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time_median",
"family_index": 1,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8611262661445145e+06,
"cpu_time": 2.0625407546174030e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2008178996287133e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.3887959968310643e+03,
"cpu_time": 2.2243965415709904e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.0076826712036977e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.8204967941762669e-03,
"cpu_time": 1.0777507638008836e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.8213261943418144e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"family_index": 1,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 379,
"real_time": 1.8658135485035826e+06,
"cpu_time": 2.0683050554090119e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3905780438619591e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"family_index": 1,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 379,
"real_time": 1.8620457329781477e+06,
"cpu_time": 2.0649793403693498e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3994622983280607e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"family_index": 1,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 379,
"real_time": 1.8668545958402352e+06,
"cpu_time": 2.0694971583113610e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3881296477259593e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"family_index": 1,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 379,
"real_time": 1.8625426178760913e+06,
"cpu_time": 2.0649337678100371e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3982886197479675e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"family_index": 1,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 379,
"real_time": 1.8574922677277867e+06,
"cpu_time": 2.0651126992084386e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4102471608245354e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8629497525851685e+06,
"cpu_time": 2.0665656042216397e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3973411540976958e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time_median",
"family_index": 1,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8625426178760913e+06,
"cpu_time": 2.0651126992084384e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3982886197479675e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6815852223612828e+03,
"cpu_time": 2.1742674556423526e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.6970320135507463e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.9762128405516246e-03,
"cpu_time": 1.0521163476255948e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.9777933321016841e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"family_index": 1,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 378,
"real_time": 1.8675173874370072e+06,
"cpu_time": 2.0719363201058132e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7731445555564575e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"family_index": 1,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 378,
"real_time": 1.8618907160806302e+06,
"cpu_time": 2.0647205608465162e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7996571756311841e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"family_index": 1,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 378,
"real_time": 1.8666060695303949e+06,
"cpu_time": 2.0698128386243146e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7774277966008782e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"family_index": 1,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 378,
"real_time": 1.8612473348896920e+06,
"cpu_time": 2.0648185978835633e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8026989712095466e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"family_index": 1,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 378,
"real_time": 1.8561980243402389e+06,
"cpu_time": 2.0650354603174550e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8266444555792902e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8626919064555925e+06,
"cpu_time": 2.0672647555555322e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7959145909154713e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time_median",
"family_index": 1,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8618907160806297e+06,
"cpu_time": 2.0650354603174548e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7996571756311841e+06,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.5690572668125587e+03,
"cpu_time": 3.3816643093493249e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.1590480615832901e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.4529323668489823e-03,
"cpu_time": 1.6358157803742831e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.4546032584413465e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"family_index": 1,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 378,
"real_time": 1.8695923067247753e+06,
"cpu_time": 2.0757830873015665e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7526815810129352e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"family_index": 1,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 378,
"real_time": 1.8635385504961642e+06,
"cpu_time": 2.0697154338624806e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7583752153276127e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"family_index": 1,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 378,
"real_time": 1.8677776931360348e+06,
"cpu_time": 2.0738081666666272e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7543843745655779e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"family_index": 1,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 378,
"real_time": 1.8641705015811592e+06,
"cpu_time": 2.0692127169311829e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7577791286905739e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"family_index": 1,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 378,
"real_time": 1.8592921918387113e+06,
"cpu_time": 2.0697394708994573e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7623910939783335e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8648742487553689e+06,
"cpu_time": 2.0716517751322631e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7571222787150066e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time_median",
"family_index": 1,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8641705015811590e+06,
"cpu_time": 2.0697394708994571e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7577791286905739e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.0040220285959804e+03,
"cpu_time": 2.9611270906739401e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.7741764511446730e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.1470734722559954e-03,
"cpu_time": 1.4293556118932628e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.1479304524581804e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"family_index": 1,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 378,
"real_time": 1.8653586467691986e+06,
"cpu_time": 2.0760680449735860e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5133190131296396e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"family_index": 1,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 378,
"real_time": 1.8612481623662577e+06,
"cpu_time": 2.0761807169312381e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5210780230768487e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"family_index": 1,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 378,
"real_time": 1.8720705885489078e+06,
"cpu_time": 2.0821512566137863e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5007226971499354e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"family_index": 1,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 378,
"real_time": 1.8659059450901286e+06,
"cpu_time": 2.0759388835978864e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5122885037399046e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"family_index": 1,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 378,
"real_time": 1.8614919658469423e+06,
"cpu_time": 2.0772519576719368e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5206168601529479e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8652150617242870e+06,
"cpu_time": 2.0775181719576865e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5136050194498554e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time_median",
"family_index": 1,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8653586467691984e+06,
"cpu_time": 2.0761807169312381e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5133190131296396e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.3901571145255748e+03,
"cpu_time": 2.6420706564912762e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.2580405412694599e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.3537002271829823e-03,
"cpu_time": 1.2717437046539048e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.3503041735073761e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"family_index": 1,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 377,
"real_time": 1.8664656728981633e+06,
"cpu_time": 2.0897463952254374e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0224704318551630e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"family_index": 1,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 377,
"real_time": 1.8700179552834786e+06,
"cpu_time": 2.0897537559681770e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0091305610020518e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"family_index": 1,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 377,
"real_time": 1.8664126735877257e+06,
"cpu_time": 2.0895803023872476e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0226698443943739e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"family_index": 1,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 377,
"real_time": 1.8751188918972260e+06,
"cpu_time": 2.0951757082228304e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 6.9900634336515427e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"family_index": 1,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 377,
"real_time": 1.8659356270672195e+06,
"cpu_time": 2.0890979893899495e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0244652655039415e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8687901641467628e+06,
"cpu_time": 2.0906708302387286e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0137599072814152e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time_median",
"family_index": 1,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8664656728981629e+06,
"cpu_time": 2.0897463952254374e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0224704318551630e+07,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.8975751717399280e+03,
"cpu_time": 2.5324202993816457e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4601529216147191e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.0856141296738107e-03,
"cpu_time": 1.2112955625312258e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.0818404691880666e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"family_index": 1,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 376,
"real_time": 1.8701654608640696e+06,
"cpu_time": 2.1090598829787755e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4017155459543243e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"family_index": 1,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 376,
"real_time": 1.8690748361513352e+06,
"cpu_time": 2.1086644361702004e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4025334616338217e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"family_index": 1,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 376,
"real_time": 1.8705612411066156e+06,
"cpu_time": 2.1088341170212440e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4014189658121902e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"family_index": 1,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 376,
"real_time": 1.8696978251992210e+06,
"cpu_time": 2.1087891648936635e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4020661331842107e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"family_index": 1,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 376,
"real_time": 1.8729858687599536e+06,
"cpu_time": 2.1092100319148973e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.3996047934603879e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8704970464162391e+06,
"cpu_time": 2.1089115265957564e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4014677800089869e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time_median",
"family_index": 1,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8701654608640694e+06,
"cpu_time": 2.1088341170212440e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4017155459543243e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4974764761038866e+03,
"cpu_time": 2.1973469759289509e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.1211816891119833e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 8.0057676593126015e-04,
"cpu_time": 1.0419341675636573e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.0000532663319142e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"family_index": 1,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 375,
"real_time": 1.8914088940558336e+06,
"cpu_time": 2.1587004799999553e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7719442456239361e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"family_index": 1,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 375,
"real_time": 1.8907149481431891e+06,
"cpu_time": 2.1587115680000200e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7729616276366067e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"family_index": 1,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 375,
"real_time": 1.8940728550466397e+06,
"cpu_time": 2.1613727999999807e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7680455828458083e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"family_index": 1,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 375,
"real_time": 1.8959424606679627e+06,
"cpu_time": 2.1609918453333043e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7653159886259794e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"family_index": 1,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 375,
"real_time": 1.8944312629755586e+06,
"cpu_time": 2.1600144400000167e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7675218956031567e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8933140841778372e+06,
"cpu_time": 2.1599582266666559e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7691578680670977e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time_median",
"family_index": 1,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8940728550466399e+06,
"cpu_time": 2.1600144400000167e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7680455828458083e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.1861748669319268e+03,
"cpu_time": 1.2458561019575714e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.1978941780490486e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.1546815635089214e-03,
"cpu_time": 5.7679638734506105e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.1548255211181635e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"family_index": 1,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 369,
"real_time": 1.8997876210041030e+06,
"cpu_time": 2.2289299566395353e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5194380066851449e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"family_index": 1,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 369,
"real_time": 1.9070053827093262e+06,
"cpu_time": 2.2340098428183720e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.4985476680210733e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"family_index": 1,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 369,
"real_time": 1.8994467825752487e+06,
"cpu_time": 2.2292730135501060e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5204284195756841e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"family_index": 1,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 369,
"real_time": 1.9062721741677271e+06,
"cpu_time": 2.2334628238482689e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5006625717432261e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"family_index": 1,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 369,
"real_time": 1.8999207456316042e+06,
"cpu_time": 2.2301946883468796e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5190512678538835e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9024865412176021e+06,
"cpu_time": 2.2311740650406326e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5116255867758024e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time_median",
"family_index": 1,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8999207456316047e+06,
"cpu_time": 2.2301946883468800e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5190512678538835e+08,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.8032405647663413e+03,
"cpu_time": 2.3921355772161587e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.1010042769762522e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.9990893403809555e-03,
"cpu_time": 1.0721420684730819e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.9976035375442094e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"family_index": 1,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 360,
"real_time": 1.9502147771870822e+06,
"cpu_time": 2.4105063972221785e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0753441233918116e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"family_index": 1,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 360,
"real_time": 1.9480943243252111e+06,
"cpu_time": 2.4085992916666674e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0765146090790136e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"family_index": 1,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 360,
"real_time": 1.9496294475781422e+06,
"cpu_time": 2.4095967444444629e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0756669697439749e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"family_index": 1,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 360,
"real_time": 1.9498054246974385e+06,
"cpu_time": 2.4091526638888656e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0755698868390553e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"family_index": 1,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 360,
"real_time": 1.9499205250111925e+06,
"cpu_time": 2.4096392972222390e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0755063978763762e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9495328997598130e+06,
"cpu_time": 2.4094988788888827e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0757203973860462e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time_median",
"family_index": 1,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9498054246974383e+06,
"cpu_time": 2.4095967444444629e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0755698868390553e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 8.3191056727917351e+02,
"cpu_time": 7.0240838749577017e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.5925430375773291e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 4.2672302036127045e-04,
"cpu_time": 2.9151637863375105e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.2692720605995844e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"family_index": 1,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 349,
"real_time": 2.0101800228050111e+06,
"cpu_time": 2.7194888796561719e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0865315307169633e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"family_index": 1,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 349,
"real_time": 2.0081514147721382e+06,
"cpu_time": 2.7180712349570254e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0886393173076153e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"family_index": 1,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 349,
"real_time": 2.0095606859543903e+06,
"cpu_time": 2.7179635787964952e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0871745896083853e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"family_index": 1,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 349,
"real_time": 2.0084356119253146e+06,
"cpu_time": 2.7178388767908197e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0883437711897974e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"family_index": 1,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 349,
"real_time": 2.0078910090413510e+06,
"cpu_time": 2.7180354040114642e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0889101953808398e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0088437488996412e+06,
"cpu_time": 2.7182795948423953e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0879198808407204e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time_median",
"family_index": 1,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0084356119253146e+06,
"cpu_time": 2.7180354040114642e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0883437711897974e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 9.8148385911700484e+02,
"cpu_time": 6.8181949602716281e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.0199227704096031e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 4.8858148357960624e-04,
"cpu_time": 2.5082758128370321e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.8848750364833052e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"family_index": 1,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 329,
"real_time": 2.1438795229629544e+06,
"cpu_time": 3.3838003069908815e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9128168864668770e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"family_index": 1,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 329,
"real_time": 2.1325885371296559e+06,
"cpu_time": 3.3817152097264286e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9335332878092813e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"family_index": 1,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 329,
"real_time": 2.1440682261842545e+06,
"cpu_time": 3.3835034285714314e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9124725125603862e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"family_index": 1,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 329,
"real_time": 2.1262551664224737e+06,
"cpu_time": 3.3752654164134283e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9452499081350780e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"family_index": 1,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 329,
"real_time": 2.1438201667180448e+06,
"cpu_time": 3.3838147750759753e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9129252211681752e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.1381223238834767e+06,
"cpu_time": 3.3816198273556293e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9233995632279596e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time_median",
"family_index": 1,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.1438201667180448e+06,
"cpu_time": 3.3835034285714314e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9129252211681752e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 8.2525258317757580e+03,
"cpu_time": 3.6574539193023475e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.5175928898768602e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 3.8597070614681565e-03,
"cpu_time": 1.0815686286540424e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.8680559178843037e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"family_index": 1,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 212,
"real_time": 3.3054854534127098e+06,
"cpu_time": 5.6353435896226671e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 5.0755679419731102e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"family_index": 1,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 212,
"real_time": 3.3067156723261159e+06,
"cpu_time": 5.6373989528300976e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 5.0736796454586105e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"family_index": 1,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 212,
"real_time": 3.3043812485938929e+06,
"cpu_time": 5.6344863207547162e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 5.0772640133880367e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"family_index": 1,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 212,
"real_time": 3.3054009007448154e+06,
"cpu_time": 5.6345207641509436e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 5.0756977757885714e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"family_index": 1,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 212,
"real_time": 3.3054771176363640e+06,
"cpu_time": 5.6359466556603294e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 5.0755807415774298e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.3054920785427801e+06,
"cpu_time": 5.6355392566037504e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 5.0755580236371527e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time_median",
"family_index": 1,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.3054771176363644e+06,
"cpu_time": 5.6353435896226671e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 5.0755807415774298e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 8.2760286966532863e+02,
"cpu_time": 1.2050653914210720e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2707134525139804e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.5037206261591642e-04,
"cpu_time": 2.1383319972600155e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.5035935883230926e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"family_index": 1,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 104,
"real_time": 6.7819850429749256e+06,
"cpu_time": 1.1429729798076838e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.9475827191269217e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"family_index": 1,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 104,
"real_time": 6.7791265587752257e+06,
"cpu_time": 1.1419085221153963e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.9496689151739674e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"family_index": 1,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 104,
"real_time": 6.7812743363901973e+06,
"cpu_time": 1.1429057490384655e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.9481012469791441e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"family_index": 1,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 104,
"real_time": 6.7853374281325024e+06,
"cpu_time": 1.1438797586538276e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.9451383008427687e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"family_index": 1,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 104,
"real_time": 6.7880927638795516e+06,
"cpu_time": 1.1428053778846156e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.9431310335869465e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 6.7831632260304810e+06,
"cpu_time": 1.1428944774999978e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.9467244431419497e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time_median",
"family_index": 1,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 6.7819850429749265e+06,
"cpu_time": 1.1429057490384657e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.9475827191269217e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.5453561209883846e+03,
"cpu_time": 6.9948469865227507e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.5850813162916172e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 5.2267002913670600e-04,
"cpu_time": 6.1202911766828141e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.2258445886864142e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"family_index": 1,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 52,
"real_time": 1.3482800863969784e+07,
"cpu_time": 2.2255891057692606e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.9773681801780262e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"family_index": 1,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 52,
"real_time": 1.3528231513471557e+07,
"cpu_time": 2.2305060230769195e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.9606531299506721e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"family_index": 1,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 52,
"real_time": 1.3475610194011377e+07,
"cpu_time": 2.2248452653846364e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.9800241349978714e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"family_index": 1,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 52,
"real_time": 1.3553585481041899e+07,
"cpu_time": 2.2314305865384981e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.9513735014154472e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"family_index": 1,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 52,
"real_time": 1.3487991805259999e+07,
"cpu_time": 2.2257048403846424e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.9754526076913195e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.3505643971550923e+07,
"cpu_time": 2.2276151642307915e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.9689743108466673e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time_median",
"family_index": 1,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.3487991805259999e+07,
"cpu_time": 2.2257048403846424e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.9754526076913195e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.3704672206723655e+04,
"cpu_time": 3.0960096074103443e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2385016454203038e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.4955990456820227e-03,
"cpu_time": 1.3898314471563647e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.4924694070500731e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"family_index": 1,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 27,
"real_time": 2.2060151453371402e+07,
"cpu_time": 4.1530629962962896e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 6.0841707403367720e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"family_index": 1,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 27,
"real_time": 2.6069903470299862e+07,
"cpu_time": 4.2372402925926268e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 5.1483784031999788e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"family_index": 1,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 27,
"real_time": 2.6175663565044049e+07,
"cpu_time": 4.2445329777777903e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 5.1275769061778183e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"family_index": 1,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 27,
"real_time": 2.6075118048875421e+07,
"cpu_time": 4.2419741370370850e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 5.1473488153887224e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"family_index": 1,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 27,
"real_time": 2.6072201905427154e+07,
"cpu_time": 4.2397978666666925e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 5.1479245399699602e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.5290607688603576e+07,
"cpu_time": 4.2233216540740967e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 5.3310798810146503e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time_median",
"family_index": 1,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.6072201905427154e+07,
"cpu_time": 4.2397978666666925e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 5.1479245399699602e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8064342869785707e+06,
"cpu_time": 3.9367843623634643e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.2108256808697420e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 7.1427081121209435e-02,
"cpu_time": 9.3215357124546289e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.8986355013466930e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"family_index": 1,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 16,
"real_time": 5.1166998222470284e+07,
"cpu_time": 8.2126040749999523e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.2462615616586046e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"family_index": 1,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 16,
"real_time": 5.1072767702862620e+07,
"cpu_time": 8.2022782562498614e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.2559410439969997e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"family_index": 1,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 16,
"real_time": 5.1108367741107941e+07,
"cpu_time": 8.2091528375000775e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.2522799663603735e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"family_index": 1,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 16,
"real_time": 4.2692180024459958e+07,
"cpu_time": 8.0069442124999672e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 6.2876961505878410e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"family_index": 1,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 16,
"real_time": 5.1081407582387328e+07,
"cpu_time": 8.2030016750000104e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.2550520571902857e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.9424344254657634e+07,
"cpu_time": 8.1667962112499744e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.4594461559588213e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time_median",
"family_index": 1,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.1081407582387328e+07,
"cpu_time": 8.2030016750000089e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.2550520571902857e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.7635744948618300e+06,
"cpu_time": 8.9463563526848820e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.6302128276372194e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 7.6148192790785690e-02,
"cpu_time": 1.0954548296871966e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.4811035686898045e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"family_index": 1,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 8,
"real_time": 9.4352710992097855e+07,
"cpu_time": 1.5206052137499881e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.6900422505609140e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"family_index": 1,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 8,
"real_time": 1.0054269246757030e+08,
"cpu_time": 1.5827201974999738e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.3397308031428127e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"family_index": 1,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 8,
"real_time": 9.3814225867390633e+07,
"cpu_time": 1.5154686512499893e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.7227025756081381e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"family_index": 1,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 8,
"real_time": 1.0046921670436859e+08,
"cpu_time": 1.5813101150000009e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.3436358877938366e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"family_index": 1,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 8,
"real_time": 9.3780806288123131e+07,
"cpu_time": 1.5166469637499702e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.7247419088141499e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 9.6591930463910118e+07,
"cpu_time": 1.5433502282499844e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.5641706851839714e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time_median",
"family_index": 1,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 9.4352710992097855e+07,
"cpu_time": 1.5206052137499881e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.6900422505609140e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.5802944990661601e+06,
"cpu_time": 3.5350848067585630e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.0357307448462942e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 3.7066186397463861e-02,
"cpu_time": 2.2905266361783739e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.6586417995173086e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"family_index": 1,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 4,
"real_time": 1.7772912979125977e+08,
"cpu_time": 2.8911530974999523e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 6.0414509723931799e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"family_index": 1,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 4,
"real_time": 1.7718904837965965e+08,
"cpu_time": 2.8867892600000286e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 6.0598656283728867e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"family_index": 1,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 4,
"real_time": 1.7743757367134094e+08,
"cpu_time": 2.8902589950000387e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 6.0513779679429131e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"family_index": 1,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 4,
"real_time": 1.7765997350215912e+08,
"cpu_time": 2.8935291049999988e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 6.0438026801065054e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"family_index": 1,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 4,
"real_time": 1.7769000679254532e+08,
"cpu_time": 2.8923193600000727e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 6.0427811523109636e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time_mean",
"family_index": 1,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.7754114642739296e+08,
"cpu_time": 2.8908099635000187e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 6.0478556802252903e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time_median",
"family_index": 1,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.7765997350215912e+08,
"cpu_time": 2.8911530974999523e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 6.0438026801065054e+09,
"dst_gpu": 1.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time_stddev",
"family_index": 1,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.2711086395882582e+05,
"cpu_time": 2.5621949650587747e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.7434374017124977e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time_cv",
"family_index": 1,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/1/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.2792012923703005e-03,
"cpu_time": 8.8632424732500343e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2803608106971304e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"family_index": 2,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 379,
"real_time": 1.8584787737374504e+06,
"cpu_time": 2.0681499340369406e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2039530705873142e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"family_index": 2,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 379,
"real_time": 1.8538442486559232e+06,
"cpu_time": 2.0683672796833452e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2094628515689429e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"family_index": 2,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 379,
"real_time": 1.8541275399278158e+06,
"cpu_time": 2.0689565092347863e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2091252687824615e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"family_index": 2,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 379,
"real_time": 1.8590964044766480e+06,
"cpu_time": 2.0687736358837958e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2032208712452757e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"family_index": 2,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 379,
"real_time": 1.8537763201428570e+06,
"cpu_time": 2.0684438364117155e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2095438136162786e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8558646573881388e+06,
"cpu_time": 2.0685382390501169e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2070611751600550e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time_median",
"family_index": 2,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8541275399278160e+06,
"cpu_time": 2.0684438364117157e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2091252687824615e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.6804185538205156e+03,
"cpu_time": 3.2375969580315467e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.1859172127235393e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.4442963516492721e-03,
"cpu_time": 1.5651617634674560e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4435110583160423e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"family_index": 2,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 379,
"real_time": 1.8578662111922745e+06,
"cpu_time": 2.0690574432718842e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4093594849022161e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"family_index": 2,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 379,
"real_time": 1.8534626635929537e+06,
"cpu_time": 2.0699982295514739e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4198354576615728e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"family_index": 2,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 379,
"real_time": 1.8580165038334997e+06,
"cpu_time": 2.0688475751978515e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4090028173044156e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"family_index": 2,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 379,
"real_time": 1.8631745187455413e+06,
"cpu_time": 2.0741065461740405e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.3967969278130746e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"family_index": 2,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 379,
"real_time": 1.8537704078801586e+06,
"cpu_time": 2.0697280633245078e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4191017211067649e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8572580610488858e+06,
"cpu_time": 2.0703475715039517e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4108192817576090e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time_median",
"family_index": 2,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8578662111922745e+06,
"cpu_time": 2.0697280633245080e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4093594849022161e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.9534263683888116e+03,
"cpu_time": 2.1534937038746893e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 9.3801277924313890e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.1286359991115698e-03,
"cpu_time": 1.0401604704036909e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.1266180256411743e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"family_index": 2,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 379,
"real_time": 1.8598243468923317e+06,
"cpu_time": 2.0726901160949848e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8094340884271134e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"family_index": 2,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 379,
"real_time": 1.8547937996353153e+06,
"cpu_time": 2.0724264564643570e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8333269192626048e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"family_index": 2,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 379,
"real_time": 1.8596879644422315e+06,
"cpu_time": 2.0720068548811623e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8100801388548985e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"family_index": 2,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 379,
"real_time": 1.8636643150753751e+06,
"cpu_time": 2.0767091319260958e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7912827795585897e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"family_index": 2,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 379,
"real_time": 1.8542272212958434e+06,
"cpu_time": 2.0723761187335397e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8360260338265859e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8584395294682193e+06,
"cpu_time": 2.0732417356200286e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8160299919859599e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time_median",
"family_index": 2,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8596879644422315e+06,
"cpu_time": 2.0724264564643570e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8100801388548985e+06,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.9309534459480137e+03,
"cpu_time": 1.9535880050969558e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.8643742927106436e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.1151903968986445e-03,
"cpu_time": 9.4228664778095016e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.1147549343700244e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"family_index": 2,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 378,
"real_time": 1.8649235953393308e+06,
"cpu_time": 2.0793030529100560e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7570693020288438e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"family_index": 2,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 378,
"real_time": 1.8592122597852412e+06,
"cpu_time": 2.0799521825395497e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7624668634545822e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"family_index": 2,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 378,
"real_time": 1.8638760258293359e+06,
"cpu_time": 2.0790380317459684e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7580568420809962e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"family_index": 2,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 378,
"real_time": 1.8684129451189644e+06,
"cpu_time": 2.0833485052909204e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7537878917828638e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"family_index": 2,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 378,
"real_time": 1.8620164956946464e+06,
"cpu_time": 2.0771524867725067e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7598125513799772e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8636882643535037e+06,
"cpu_time": 2.0797588518518005e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7582386901454527e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time_median",
"family_index": 2,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8638760258293361e+06,
"cpu_time": 2.0793030529100560e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7580568420809962e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.4173576303641212e+03,
"cpu_time": 2.2607223288513101e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.2235319164553421e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.8336530286354360e-03,
"cpu_time": 1.0870117594827791e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.8333869767071676e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"family_index": 2,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 373,
"real_time": 1.8631497094262238e+06,
"cpu_time": 2.0841250697050730e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5174843797271930e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"family_index": 2,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 373,
"real_time": 1.8583797352413782e+06,
"cpu_time": 2.0844219517426065e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5265128411168218e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"family_index": 2,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 373,
"real_time": 1.8637386820046231e+06,
"cpu_time": 2.0840886461125186e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5163727958637409e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"family_index": 2,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 373,
"real_time": 1.8683675662032329e+06,
"cpu_time": 2.0895920589813108e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5076609755744003e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"family_index": 2,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 373,
"real_time": 1.8625787971790968e+06,
"cpu_time": 2.0833736863270456e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5185625488304302e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8632428980109110e+06,
"cpu_time": 2.0851202825737111e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5173187082225181e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time_median",
"family_index": 2,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8631497094262238e+06,
"cpu_time": 2.0841250697050733e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5174843797271930e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.5569682313689545e+03,
"cpu_time": 2.5293166605767956e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.7132930860625318e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.9090201471671597e-03,
"cpu_time": 1.2130315367009919e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.9086394048877940e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"family_index": 2,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 374,
"real_time": 1.8620722242590552e+06,
"cpu_time": 2.0974661871657632e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0390395330747917e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"family_index": 2,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 374,
"real_time": 1.8659045223385112e+06,
"cpu_time": 2.1020039278073576e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0245823637175903e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"family_index": 2,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 374,
"real_time": 1.8574399007113904e+06,
"cpu_time": 2.0968358877006173e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0565943990866169e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"family_index": 2,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 374,
"real_time": 1.8622664210463148e+06,
"cpu_time": 2.0976107112299232e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0383055033746019e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"family_index": 2,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 374,
"real_time": 1.8580696193571698e+06,
"cpu_time": 2.0968186898395997e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0542028476493001e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8611505375424884e+06,
"cpu_time": 2.0981470807486521e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0425449293805808e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time_median",
"family_index": 2,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8620722242590550e+06,
"cpu_time": 2.0974661871657632e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0390395330747917e+07,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.4625119637631310e+03,
"cpu_time": 2.1857737292887223e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.3097795221429445e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.8604147778046598e-03,
"cpu_time": 1.0417638254934937e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.8598099625587261e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"family_index": 2,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 378,
"real_time": 1.8632719163501405e+06,
"cpu_time": 2.1213788068782301e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4069014710075128e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"family_index": 2,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 378,
"real_time": 1.8626099278215349e+06,
"cpu_time": 2.1213662486771950e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4074014966010490e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"family_index": 2,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 378,
"real_time": 1.8624405988101556e+06,
"cpu_time": 2.1210991322752149e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4075294544560191e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"family_index": 2,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 378,
"real_time": 1.8621527560368776e+06,
"cpu_time": 2.1207425608466147e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4077470237076971e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"family_index": 2,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 378,
"real_time": 1.8616101074570830e+06,
"cpu_time": 2.1200421481482587e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4081573738234735e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8624170612951580e+06,
"cpu_time": 2.1209257793651028e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4075473639191505e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time_median",
"family_index": 2,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8624405988101556e+06,
"cpu_time": 2.1210991322752149e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4075294544560191e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 6.1023774971939963e+02,
"cpu_time": 5.5760010913291501e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.6118103549473934e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 3.2765902031365056e-04,
"cpu_time": 2.6290411223151434e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.2764867976494577e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"family_index": 2,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 372,
"real_time": 1.8943431925670726e+06,
"cpu_time": 2.1887299973118557e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7676505611927903e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"family_index": 2,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 372,
"real_time": 1.8956631879567581e+06,
"cpu_time": 2.1871740161289317e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7657233802440625e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"family_index": 2,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 372,
"real_time": 1.8973913543196695e+06,
"cpu_time": 2.1891521370968525e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7632043268584895e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"family_index": 2,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 372,
"real_time": 1.8929414571918827e+06,
"cpu_time": 2.1876962876345161e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7697000243090677e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"family_index": 2,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 372,
"real_time": 1.8914356637309226e+06,
"cpu_time": 2.1862199408602887e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7719050140242344e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8943549711532616e+06,
"cpu_time": 2.1877944758064887e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7676366613257289e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time_median",
"family_index": 2,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8943431925670723e+06,
"cpu_time": 2.1876962876345157e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7676505611927903e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.3157919928368551e+03,
"cpu_time": 1.1823933452441490e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.3831467635915533e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.2224699320355088e-03,
"cpu_time": 5.4044991808852700e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2223955589499194e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"family_index": 2,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 373,
"real_time": 1.8888921616567753e+06,
"cpu_time": 2.2673993083109008e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5512750875109696e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"family_index": 2,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 373,
"real_time": 1.8914024232126442e+06,
"cpu_time": 2.2685671769436486e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5439074579324043e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"family_index": 2,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 373,
"real_time": 1.8887180192383560e+06,
"cpu_time": 2.2672047024128679e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5517869227659965e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"family_index": 2,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 373,
"real_time": 1.8894155086867071e+06,
"cpu_time": 2.2659012412867285e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5497374462054825e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"family_index": 2,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 373,
"real_time": 1.8893039936540604e+06,
"cpu_time": 2.2664420643431535e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5500650161225390e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8895464212897089e+06,
"cpu_time": 2.2671028986594598e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5493543861074793e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time_median",
"family_index": 2,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8893039936540604e+06,
"cpu_time": 2.2672047024128679e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5500650161225390e+08,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.0764476332216750e+03,
"cpu_time": 1.0152605592671423e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.1593941292595956e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 5.6968573044474159e-04,
"cpu_time": 4.4782288438141333e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.6932643140776432e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"family_index": 2,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 364,
"real_time": 1.9354302766169985e+06,
"cpu_time": 2.4703439780220115e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0835585375184274e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"family_index": 2,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 364,
"real_time": 1.9352927130447435e+06,
"cpu_time": 2.4694505439559999e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0836355585200379e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"family_index": 2,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 364,
"real_time": 1.9346689583111943e+06,
"cpu_time": 2.4697988626372875e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0839849324044771e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"family_index": 2,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 364,
"real_time": 1.9340197124600329e+06,
"cpu_time": 2.4700066978021800e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0843488235869460e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"family_index": 2,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 364,
"real_time": 1.9357617156756632e+06,
"cpu_time": 2.4710101978021506e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0833730117800190e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9350346752217268e+06,
"cpu_time": 2.4701220560439257e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0837801727619815e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time_median",
"family_index": 2,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9352927130447433e+06,
"cpu_time": 2.4700066978021795e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0836355585200379e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 6.9202060482630736e+02,
"cpu_time": 5.9302932667800303e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.8765506440649013e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 3.5762697882765949e-04,
"cpu_time": 2.4008098111061814e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.5768790954955631e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"family_index": 2,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 344,
"real_time": 2.0317115403338186e+06,
"cpu_time": 2.8874096482558195e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0644190460772099e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"family_index": 2,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 344,
"real_time": 2.0318385203461009e+06,
"cpu_time": 2.8865069331395598e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0642900299407392e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"family_index": 2,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 344,
"real_time": 2.0305585247453649e+06,
"cpu_time": 2.8857650494186333e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0655912887445447e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"family_index": 2,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 344,
"real_time": 2.0300031766203386e+06,
"cpu_time": 2.8862562034883169e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0661563727121401e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"family_index": 2,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 344,
"real_time": 2.0282682857806468e+06,
"cpu_time": 2.8856518720929306e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0679236713429563e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0304760095652540e+06,
"cpu_time": 2.8863179412790518e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0656760817635183e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time_median",
"family_index": 2,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0305585247453649e+06,
"cpu_time": 2.8862562034883173e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0655912887445447e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4566567857917748e+03,
"cpu_time": 7.0368515156399121e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4824647159106352e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 7.1739669857200637e-04,
"cpu_time": 2.4380029015519955e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.1766562482778937e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"family_index": 2,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 337,
"real_time": 2.0770264726590905e+06,
"cpu_time": 3.6424380326410378e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0387583453669586e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"family_index": 2,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 337,
"real_time": 2.0782841899423953e+06,
"cpu_time": 3.6422785608308478e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0363142060145831e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"family_index": 2,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 337,
"real_time": 2.0729624182649110e+06,
"cpu_time": 3.6424999881305816e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0466763536510916e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"family_index": 2,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 337,
"real_time": 2.0772021330163458e+06,
"cpu_time": 3.6450767181008817e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0384168043476534e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"family_index": 2,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 337,
"real_time": 2.0805212792744227e+06,
"cpu_time": 3.6433564569733115e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0319741420407438e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0771992986314329e+06,
"cpu_time": 3.6431299513353324e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0384279702842064e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time_median",
"family_index": 2,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0772021330163456e+06,
"cpu_time": 3.6424999881305820e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0384168043476534e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.7474341047905868e+03,
"cpu_time": 1.1663700701148286e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.3447403800895698e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.3226627346739138e-03,
"cpu_time": 3.2015604320875622e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.3234705235347880e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"family_index": 2,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 189,
"real_time": 3.7013458607905600e+06,
"cpu_time": 6.8072491428569425e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5327339381401672e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"family_index": 2,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 189,
"real_time": 3.7068934303033287e+06,
"cpu_time": 6.8089752962964904e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5259504529719238e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"family_index": 2,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 189,
"real_time": 3.7079922788377320e+06,
"cpu_time": 6.8133620952381827e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5246092058365364e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"family_index": 2,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 189,
"real_time": 3.7050131713311193e+06,
"cpu_time": 6.7998078253967874e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5282473298124237e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"family_index": 2,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 189,
"real_time": 3.7009827669453686e+06,
"cpu_time": 6.7964216878308635e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5331786329411068e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.7044455016416223e+06,
"cpu_time": 6.8051632095238538e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5289439119404316e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time_median",
"family_index": 2,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.7050131713311188e+06,
"cpu_time": 6.8072491428569434e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5282473298124237e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.1816873297865886e+03,
"cpu_time": 6.9137397806147319e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.8901433394259396e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 8.5888355716844187e-04,
"cpu_time": 1.0159550282260572e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.5895153816537397e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"family_index": 2,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 95,
"real_time": 7.5182302511836356e+06,
"cpu_time": 1.3941376136842553e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4630758674512987e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"family_index": 2,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 95,
"real_time": 7.5192625732406192e+06,
"cpu_time": 1.3943871778947499e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4624631302825823e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"family_index": 2,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 95,
"real_time": 7.5243641034160787e+06,
"cpu_time": 1.3948730052631594e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4594375735706635e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"family_index": 2,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 95,
"real_time": 7.5245157227312261e+06,
"cpu_time": 1.3937407968421068e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4593477157119303e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"family_index": 2,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 95,
"real_time": 7.5229561544562643e+06,
"cpu_time": 1.3933047147368502e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4602721737416811e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.5218657610055655e+06,
"cpu_time": 1.3940886616842244e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4609192921516314e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time_median",
"family_index": 2,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.5229561544562634e+06,
"cpu_time": 1.3941376136842553e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4602721737416811e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.9345565258432589e+03,
"cpu_time": 6.0047602786422967e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.7405715938392193e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 3.9013678508548004e-04,
"cpu_time": 4.3073015681713062e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.9018226509982229e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"family_index": 2,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 48,
"real_time": 1.4020963921211660e+07,
"cpu_time": 2.5905443437500492e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.7863231356350718e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"family_index": 2,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 48,
"real_time": 1.4682143072908124e+07,
"cpu_time": 2.7087393875000030e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5707812317829161e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"family_index": 2,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 48,
"real_time": 1.4732850172246495e+07,
"cpu_time": 2.7148618749999780e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5550496486021824e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"family_index": 2,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 48,
"real_time": 1.4725653377051154e+07,
"cpu_time": 2.7149862104166072e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5572758153186073e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"family_index": 2,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 48,
"real_time": 1.4715023299989602e+07,
"cpu_time": 2.7134790291666400e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5605679740953875e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4575326768681407e+07,
"cpu_time": 2.6885221691666555e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6059995610868330e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time_median",
"family_index": 2,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4715023299989602e+07,
"cpu_time": 2.7134790291666400e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5605679740953875e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.1050552419302322e+05,
"cpu_time": 5.4830102667761955e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.0098372187791535e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.1303503456280581e-02,
"cpu_time": 2.0394141918032724e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.1924388080941805e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"family_index": 2,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 25,
"real_time": 2.8413940295577049e+07,
"cpu_time": 5.1241220520000756e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7236577047672796e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"family_index": 2,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 25,
"real_time": 2.8384627923369408e+07,
"cpu_time": 5.1223832319999479e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7285357540127172e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"family_index": 2,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 25,
"real_time": 2.8419796675443649e+07,
"cpu_time": 5.1314089239999704e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7226843151897669e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"family_index": 2,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 25,
"real_time": 2.5732265785336494e+07,
"cpu_time": 4.8830315239999898e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 5.2159312016932392e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"family_index": 2,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 25,
"real_time": 2.8364084139466286e+07,
"cpu_time": 5.1185832559999660e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7319605787393322e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.7862942963838577e+07,
"cpu_time": 5.0759057975999899e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.8245539108804674e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time_median",
"family_index": 2,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.8384627923369408e+07,
"cpu_time": 5.1223832319999471e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7285357540127172e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.1912983494502562e+06,
"cpu_time": 1.0792058076325529e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.1881879164244372e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 4.2755654023925657e-02,
"cpu_time": 2.1261344293324498e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.5355238159730693e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"family_index": 2,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 10,
"real_time": 5.5166532099246979e+07,
"cpu_time": 9.5859054400000334e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8659113738937407e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"family_index": 2,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 10,
"real_time": 5.5307603627443314e+07,
"cpu_time": 9.5938668899998456e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8535000324404564e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"family_index": 2,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 10,
"real_time": 5.2567259967327118e+07,
"cpu_time": 9.2994377800005168e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.1065141338324375e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"family_index": 2,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 10,
"real_time": 5.5148562788963318e+07,
"cpu_time": 9.5761200800001234e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8674968562140121e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"family_index": 2,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 10,
"real_time": 5.5126883462071419e+07,
"cpu_time": 9.5581788200001940e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8694110593915558e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.4663368389010444e+07,
"cpu_time": 9.5227018020001441e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9125666911544409e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time_median",
"family_index": 2,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.5148562788963318e+07,
"cpu_time": 9.5761200800001234e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8674968562140121e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.1738976856912039e+06,
"cpu_time": 1.2551705497316974e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.0859891562082855e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.1475033835039063e-02,
"cpu_time": 1.3180823844216797e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.2106349378700867e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"family_index": 2,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 8,
"real_time": 1.0772499628365040e+08,
"cpu_time": 1.8389215724999985e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.9837171549894199e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"family_index": 2,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 8,
"real_time": 9.8221935331821442e+07,
"cpu_time": 1.7470415225000125e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.4658962907450199e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"family_index": 2,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 8,
"real_time": 1.0784747824072838e+08,
"cpu_time": 1.8433919000000286e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.9780571670080252e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"family_index": 2,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 8,
"real_time": 9.8238898441195488e+07,
"cpu_time": 1.7494883512500080e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.4649524833726006e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"family_index": 2,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 8,
"real_time": 1.0750733315944672e+08,
"cpu_time": 1.8381612712499872e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.9938073638544617e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.0390812829136848e+08,
"cpu_time": 1.8034009235000068e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.1772860919939060e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time_median",
"family_index": 2,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.0750733315944672e+08,
"cpu_time": 1.8381612712499875e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.9938073638544617e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.1844525372433672e+06,
"cpu_time": 5.0379122685382264e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.6309377660624054e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 4.9894581131378464e-02,
"cpu_time": 2.7935619877363378e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.0816928392867000e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"family_index": 2,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 4,
"real_time": 1.9372923299670219e+08,
"cpu_time": 3.1309286350000322e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.5424873540808268e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"family_index": 2,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 4,
"real_time": 1.9393130391836166e+08,
"cpu_time": 3.1182908949999446e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.5367122393608408e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"family_index": 2,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 4,
"real_time": 1.9423726946115494e+08,
"cpu_time": 3.1378722150000727e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.5279907248425112e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"family_index": 2,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 4,
"real_time": 1.9410534948110580e+08,
"cpu_time": 3.1333693674999094e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.5317477177748671e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"family_index": 2,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 4,
"real_time": 1.9406627118587494e+08,
"cpu_time": 3.1344527999999624e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.5328616221598845e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time_mean",
"family_index": 2,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9401388540863994e+08,
"cpu_time": 3.1309827824999845e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.5343599316437874e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time_median",
"family_index": 2,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9406627118587494e+08,
"cpu_time": 3.1333693674999100e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.5328616221598845e+09,
"dst_gpu": 2.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time_stddev",
"family_index": 2,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9291000444967538e+05,
"cpu_time": 7.5214365289334487e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.5050825349380551e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time_cv",
"family_index": 2,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/2/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 9.9431029920028931e-04,
"cpu_time": 2.4022605844315230e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 9.9470988568373868e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"family_index": 3,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 380,
"real_time": 1.8537589148955273e+06,
"cpu_time": 2.0660827131579434e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2095645593865365e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"family_index": 3,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 380,
"real_time": 1.8493217583837626e+06,
"cpu_time": 2.0662566736841896e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2148660618040580e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"family_index": 3,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 380,
"real_time": 1.8540018368137106e+06,
"cpu_time": 2.0658586078947170e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2092750496080359e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"family_index": 3,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 380,
"real_time": 1.8583563830026451e+06,
"cpu_time": 2.0707554052630952e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2040982222052994e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"family_index": 3,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 380,
"real_time": 1.8492572971904576e+06,
"cpu_time": 2.0667823710525860e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2149432673446671e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8529392380572204e+06,
"cpu_time": 2.0671471542105065e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2105494320697193e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time_median",
"family_index": 3,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8537589148955271e+06,
"cpu_time": 2.0662566736841896e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2095645593865365e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.8009634741723939e+03,
"cpu_time": 2.0456590832704194e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.5318653518054280e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.0513157669204782e-03,
"cpu_time": 9.8960496310273888e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.0501081251832853e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"family_index": 3,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 380,
"real_time": 1.8536025913099844e+06,
"cpu_time": 2.0669739921053657e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4195018060535416e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"family_index": 3,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 380,
"real_time": 1.8491599716454449e+06,
"cpu_time": 2.0674041394736259e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4301196898127105e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"family_index": 3,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 380,
"real_time": 1.8544006895114619e+06,
"cpu_time": 2.0677960631578690e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4175997379283579e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"family_index": 3,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 380,
"real_time": 1.8579019931501918e+06,
"cpu_time": 2.0717456052630879e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4092745635683071e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"family_index": 3,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 380,
"real_time": 1.8540981285408826e+06,
"cpu_time": 2.0673977578947020e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4183206238640929e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8538326748315929e+06,
"cpu_time": 2.0682635115789298e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4189632842454016e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time_median",
"family_index": 3,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8540981285408824e+06,
"cpu_time": 2.0674041394736257e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4183206238640929e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.1160692955870454e+03,
"cpu_time": 1.9681460049311838e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.4313351797446830e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.6808794762828943e-03,
"cpu_time": 9.5159344731111386e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.6816919946447768e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"family_index": 3,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 376,
"real_time": 1.8552583452471115e+06,
"cpu_time": 2.0702489707448264e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8311151069463212e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"family_index": 3,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 376,
"real_time": 1.8520032796933518e+06,
"cpu_time": 2.0717846436170968e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8466366013740581e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"family_index": 3,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 376,
"real_time": 1.8572147936673842e+06,
"cpu_time": 2.0715562632979206e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8218121328050736e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"family_index": 3,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 376,
"real_time": 1.8620535573527960e+06,
"cpu_time": 2.0763282500001455e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.7988876234539952e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"family_index": 3,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 376,
"real_time": 1.8569033136760057e+06,
"cpu_time": 2.0715255132979276e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8232919179650396e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8566866579273299e+06,
"cpu_time": 2.0722887281915832e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8243486765088979e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time_median",
"family_index": 3,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8569033136760057e+06,
"cpu_time": 2.0715562632979203e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8232919179650396e+06,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6436103424936687e+03,
"cpu_time": 2.3372767238280776e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.7308611952286752e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.9624260921663169e-03,
"cpu_time": 1.1278721406103194e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.9614605663037343e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"family_index": 3,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 375,
"real_time": 1.8597307271556926e+06,
"cpu_time": 2.0769743999999263e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7619755119127378e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"family_index": 3,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 375,
"real_time": 1.8560639371668608e+06,
"cpu_time": 2.0776317680000223e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7654564233394802e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"family_index": 3,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 375,
"real_time": 1.8606267278664745e+06,
"cpu_time": 2.0772982080000779e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7611270175385524e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"family_index": 3,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 375,
"real_time": 1.8657702140044423e+06,
"cpu_time": 2.0827778266666429e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7562720078841385e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"family_index": 3,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 375,
"real_time": 1.8593578267997750e+06,
"cpu_time": 2.0768149813334274e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7623288819236308e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8603098865986492e+06,
"cpu_time": 2.0782994368000198e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7614319685197081e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time_median",
"family_index": 3,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8597307271556929e+06,
"cpu_time": 2.0772982080000781e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7619755119127378e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.5066185228974159e+03,
"cpu_time": 2.5230709010635833e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.3171643962925998e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.8849647298863966e-03,
"cpu_time": 1.2140074025850591e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.8832202750812541e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"family_index": 3,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 379,
"real_time": 1.8564926668374042e+06,
"cpu_time": 2.0847415145118490e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5300974343002774e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"family_index": 3,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 379,
"real_time": 1.8568342025723495e+06,
"cpu_time": 2.0849925672823065e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5294481278517090e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"family_index": 3,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 379,
"real_time": 1.8609663583157035e+06,
"cpu_time": 2.0842011345645655e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5216112159767561e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"family_index": 3,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 379,
"real_time": 1.8679274393036400e+06,
"cpu_time": 2.0914431424802805e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5084874616131611e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"family_index": 3,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 379,
"real_time": 1.8606969963751198e+06,
"cpu_time": 2.0843719287598401e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5221210185039625e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8605835326808435e+06,
"cpu_time": 2.0859500575197686e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5223530516491733e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time_median",
"family_index": 3,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8606969963751198e+06,
"cpu_time": 2.0847415145118490e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5221210185039625e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.6066542515263154e+03,
"cpu_time": 3.0862828785822080e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.7059507744702991e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.4759190708781363e-03,
"cpu_time": 1.4795574167541926e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.4716292338708506e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"family_index": 3,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 379,
"real_time": 1.8567763493024057e+06,
"cpu_time": 2.0990176174142361e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0591161961559877e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"family_index": 3,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 379,
"real_time": 1.8652812882644765e+06,
"cpu_time": 2.1038509630607893e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0269294408648685e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"family_index": 3,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 379,
"real_time": 1.8576168831306358e+06,
"cpu_time": 2.0987151767810420e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0559220897639975e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"family_index": 3,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 379,
"real_time": 1.8625287688106892e+06,
"cpu_time": 2.0995097255936619e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0373141180361763e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"family_index": 3,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 379,
"real_time": 1.8581935640124695e+06,
"cpu_time": 2.0993299208443491e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0537323203816891e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8600793707041356e+06,
"cpu_time": 2.1000846807388160e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0466028330405444e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time_median",
"family_index": 3,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8581935640124697e+06,
"cpu_time": 2.0993299208443491e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0537323203816891e+07,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6602445170588339e+03,
"cpu_time": 2.1271626449276328e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.3852662002301219e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.9677894259282299e-03,
"cpu_time": 1.0128937487317372e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.9658638822877893e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"family_index": 3,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 378,
"real_time": 1.8619373587416559e+06,
"cpu_time": 2.1232776111111841e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4079098782204118e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"family_index": 3,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 378,
"real_time": 1.8626671082635222e+06,
"cpu_time": 2.1229460846560868e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4073582919729796e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"family_index": 3,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 378,
"real_time": 1.8626425409733895e+06,
"cpu_time": 2.1229382698412077e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4073768542997381e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"family_index": 3,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 378,
"real_time": 1.8666429570186194e+06,
"cpu_time": 2.1275644629630428e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4043606947666809e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"family_index": 3,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 378,
"real_time": 1.8617731113514677e+06,
"cpu_time": 2.1224437962963469e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4080340853655833e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8631326152697310e+06,
"cpu_time": 2.1238340449735736e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4070079609250790e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time_median",
"family_index": 3,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8626425409733895e+06,
"cpu_time": 2.1229460846560863e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4073768542997381e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0035130305030302e+03,
"cpu_time": 2.1064967965250898e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.5110607572165984e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.0753464429116744e-03,
"cpu_time": 9.9183681583336751e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.0739532392006559e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"family_index": 3,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 372,
"real_time": 1.8937612504457994e+06,
"cpu_time": 2.1918054489246886e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7685010445565951e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"family_index": 3,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 372,
"real_time": 1.8902137453551404e+06,
"cpu_time": 2.1879094811827792e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7736968969162524e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"family_index": 3,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 372,
"real_time": 1.8911814195528296e+06,
"cpu_time": 2.1882802580645122e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7722776597708327e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"family_index": 3,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 372,
"real_time": 1.8900717639612155e+06,
"cpu_time": 2.1879663629032555e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7739052558575678e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"family_index": 3,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 372,
"real_time": 1.8942859933695316e+06,
"cpu_time": 2.1921322715053475e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7677341322014594e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8919028345369033e+06,
"cpu_time": 2.1896187645161166e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7712229978605413e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time_median",
"family_index": 3,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8911814195528298e+06,
"cpu_time": 2.1882802580645117e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7722776597708327e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9911954327784679e+03,
"cpu_time": 2.1530783897929423e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.9158279109028366e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.0524829269394638e-03,
"cpu_time": 9.8331199233614116e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.0521809010512449e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"family_index": 3,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 373,
"real_time": 1.8894862804634261e+06,
"cpu_time": 2.2721266702413126e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5495295776522934e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"family_index": 3,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 373,
"real_time": 1.8944073019483082e+06,
"cpu_time": 2.2725137560320981e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5351138000871778e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"family_index": 3,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 373,
"real_time": 1.8899058120234073e+06,
"cpu_time": 2.2716485898123863e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5482976629261398e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"family_index": 3,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 373,
"real_time": 1.8940388195212909e+06,
"cpu_time": 2.2730954879355780e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5361906482203066e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"family_index": 3,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 373,
"real_time": 1.8908469835527232e+06,
"cpu_time": 2.2727305415550787e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5455359905952024e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8917370395018314e+06,
"cpu_time": 2.2724230091152908e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5429335358962238e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time_median",
"family_index": 3,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8908469835527230e+06,
"cpu_time": 2.2725137560320981e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5455359905952024e+08,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.3259408153047311e+03,
"cpu_time": 5.5735557721405826e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.8130242635704740e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.2295264969370387e-03,
"cpu_time": 2.4526928964297464e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2291369217128619e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"family_index": 3,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 363,
"real_time": 1.9358205172866993e+06,
"cpu_time": 2.4725975426997878e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0833401037299819e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"family_index": 3,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 363,
"real_time": 1.9336964479024129e+06,
"cpu_time": 2.4709552837465438e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0845300989588599e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"family_index": 3,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 363,
"real_time": 1.9346150011914594e+06,
"cpu_time": 2.4711756143250829e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0840151651405783e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"family_index": 3,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 363,
"real_time": 1.9350681212132915e+06,
"cpu_time": 2.4727461680440959e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0837613296451194e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"family_index": 3,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 363,
"real_time": 1.9360409165057801e+06,
"cpu_time": 2.4717076088154749e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0832167761128714e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9350482008199287e+06,
"cpu_time": 2.4718364435261972e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0837726947174823e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time_median",
"family_index": 3,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9350681212132915e+06,
"cpu_time": 2.4717076088154754e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0837613296451194e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 9.4832112773509630e+02,
"cpu_time": 8.1188158329860994e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.3121622535461024e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 4.9007623031471184e-04,
"cpu_time": 3.2845279283139810e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.9015464953477863e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"family_index": 3,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 345,
"real_time": 2.0341536794465196e+06,
"cpu_time": 2.8890419478260730e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0619405713442671e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"family_index": 3,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 345,
"real_time": 2.0321436990659845e+06,
"cpu_time": 2.8870881420289152e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0639800236212573e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"family_index": 3,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 345,
"real_time": 2.0309564640637543e+06,
"cpu_time": 2.8865827304348778e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0651865631858940e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"family_index": 3,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 345,
"real_time": 2.0295377798796890e+06,
"cpu_time": 2.8878096956522432e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0666301665242412e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"family_index": 3,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 345,
"real_time": 2.0346197687710326e+06,
"cpu_time": 2.8872689449276486e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0614682233887255e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0322822782453964e+06,
"cpu_time": 2.8875582921739514e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0638411096128774e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time_median",
"family_index": 3,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0321436990659847e+06,
"cpu_time": 2.8872689449276486e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0639800236212573e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.1374788372068588e+03,
"cpu_time": 9.3820593411470600e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.1708878145496142e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.0517627694181764e-03,
"cpu_time": 3.2491324474989572e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.0518677065003400e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"family_index": 3,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 337,
"real_time": 2.0814954395784556e+06,
"cpu_time": 3.6496735964392726e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0300871385521078e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"family_index": 3,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 337,
"real_time": 2.0728957831019463e+06,
"cpu_time": 3.6469538338279747e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0468064378263259e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"family_index": 3,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 337,
"real_time": 2.0820409980483728e+06,
"cpu_time": 3.6466020801187572e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0290311323663492e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"family_index": 3,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 337,
"real_time": 2.0748831702048064e+06,
"cpu_time": 3.6465603353115134e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0429302817912307e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"family_index": 3,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 337,
"real_time": 2.0752450142426067e+06,
"cpu_time": 3.6471784094955944e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0422253480568190e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0773120810352378e+06,
"cpu_time": 3.6473936510386220e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0382160677185664e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time_median",
"family_index": 3,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0752450142426069e+06,
"cpu_time": 3.6469538338279747e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0422253480568190e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.1695085201062839e+03,
"cpu_time": 1.2998960298874031e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.1014679005720932e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.0071652007282365e-03,
"cpu_time": 3.5639038564352601e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.0061997091574907e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"family_index": 3,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 189,
"real_time": 3.6969183692856440e+06,
"cpu_time": 6.8005900264549712e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5381624164024649e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"family_index": 3,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 189,
"real_time": 3.5593321327632540e+06,
"cpu_time": 6.8502048835977716e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.7135854070957870e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"family_index": 3,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 189,
"real_time": 3.6924239987182240e+06,
"cpu_time": 6.7894136137566883e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5436862087950869e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"family_index": 3,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 189,
"real_time": 3.6963299483494470e+06,
"cpu_time": 6.7970770317460680e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5388848491438570e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"family_index": 3,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 189,
"real_time": 3.6939553960803011e+06,
"cpu_time": 6.7932117407407435e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5418025398472595e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6677919690393740e+06,
"cpu_time": 6.8060994592592493e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5752242842568913e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time_median",
"family_index": 3,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6939553960803011e+06,
"cpu_time": 6.7970770317460671e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5418025398472595e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 6.0657978329491962e+04,
"cpu_time": 2.5007784952243703e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.7378291706475914e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.6538009473143267e-02,
"cpu_time": 3.6743196454795063e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.6912458690327072e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"family_index": 3,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 95,
"real_time": 7.5076852434952008e+06,
"cpu_time": 1.3941601515789084e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4693445331997890e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"family_index": 3,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 95,
"real_time": 7.5224137178769233e+06,
"cpu_time": 1.3939104821052920e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4605938011968555e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"family_index": 3,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 95,
"real_time": 7.5168810735799763e+06,
"cpu_time": 1.3943381705263093e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4638769286819944e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"family_index": 3,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 95,
"real_time": 7.5229929858132415e+06,
"cpu_time": 1.3942468305263469e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4602503369704714e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"family_index": 3,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 95,
"real_time": 7.5131353775137346e+06,
"cpu_time": 1.3937591136842504e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4661024078477230e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.5166216796558155e+06,
"cpu_time": 1.3940829496842215e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4640336015793676e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time_median",
"family_index": 3,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.5168810735799763e+06,
"cpu_time": 1.3941601515789086e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4638769286819944e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 6.4463696756315640e+03,
"cpu_time": 2.4112733140080281e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.8293543247706918e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 8.5761528920353248e-04,
"cpu_time": 1.7296483789248077e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.5782381284403250e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"family_index": 3,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 48,
"real_time": 1.4701629639603198e+07,
"cpu_time": 2.7085812145833421e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5647227991121731e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"family_index": 3,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 48,
"real_time": 1.4686109730973840e+07,
"cpu_time": 2.7088345874999460e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5695466824998312e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"family_index": 3,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 48,
"real_time": 1.4702882986360541e+07,
"cpu_time": 2.7128969583333883e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5643336794732742e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"family_index": 3,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 48,
"real_time": 1.4692206265560040e+07,
"cpu_time": 2.7051310083333392e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5676505479854107e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"family_index": 3,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 48,
"real_time": 1.4662789568925897e+07,
"cpu_time": 2.7044841937499344e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5768142333720999e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4689123638284704e+07,
"cpu_time": 2.7079855924999900e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5686135884885578e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time_median",
"family_index": 3,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4692206265560042e+07,
"cpu_time": 2.7085812145833425e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5676505479854107e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.6261835425504201e+04,
"cpu_time": 3.3765928274303435e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.0622265634007333e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.1070664136232395e-03,
"cpu_time": 1.2469020650560775e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.1080443695557711e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"family_index": 3,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 25,
"real_time": 2.6999604105949402e+07,
"cpu_time": 4.9967930839998186e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.9710998529206190e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"family_index": 3,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 25,
"real_time": 2.8340256065130234e+07,
"cpu_time": 5.1083491759998195e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7359391422415934e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"family_index": 3,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 25,
"real_time": 2.8328038081526756e+07,
"cpu_time": 5.1020464680000260e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7379817696420670e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"family_index": 3,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 25,
"real_time": 2.8411948829889297e+07,
"cpu_time": 5.1134010199998558e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7239887979385386e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"family_index": 3,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 25,
"real_time": 2.8396064192056656e+07,
"cpu_time": 5.1183771519999936e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7266313772295694e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.8095182254910477e+07,
"cpu_time": 5.0877933799999028e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7791281879944773e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time_median",
"family_index": 3,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.8340256065130241e+07,
"cpu_time": 5.1083491759998202e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7359391422415934e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 6.1348298510445107e+05,
"cpu_time": 5.1229540977157111e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.0747989233691667e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.1835878462658009e-02,
"cpu_time": 1.0069107990615391e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.2489434915538380e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"family_index": 3,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 10,
"real_time": 5.0316881388425827e+07,
"cpu_time": 9.1693882499998838e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.3348985190037441e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"family_index": 3,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 10,
"real_time": 5.5149123072624207e+07,
"cpu_time": 9.5626295699997857e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8674474052199430e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"family_index": 3,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 10,
"real_time": 5.5211555957794189e+07,
"cpu_time": 9.5822959800000265e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8619433258718929e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"family_index": 3,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 10,
"real_time": 5.5132771283388138e+07,
"cpu_time": 9.5681858000000373e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8688910379674921e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"family_index": 3,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 10,
"real_time": 5.0012272223830223e+07,
"cpu_time": 9.1286271699999586e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.3673917233477316e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.3164520785212524e+07,
"cpu_time": 9.4022253539999396e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.0601144022821608e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time_median",
"family_index": 3,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.5132771283388138e+07,
"cpu_time": 9.5626295699997857e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8688910379674921e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.7408361127319322e+06,
"cpu_time": 2.3171474693160695e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.6593438344171444e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 5.1553857201216108e-02,
"cpu_time": 2.4644670618645590e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.2555014037187667e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"family_index": 3,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 8,
"real_time": 9.5046305097639561e+07,
"cpu_time": 1.8012250099999961e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.6485195447469635e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"family_index": 3,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 8,
"real_time": 1.0783355310559273e+08,
"cpu_time": 1.8340230550000313e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.9787000106941252e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"family_index": 3,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 8,
"real_time": 9.4974483363330364e+07,
"cpu_time": 1.7999204125000289e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.6527910759584694e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"family_index": 3,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 8,
"real_time": 1.0786197148263454e+08,
"cpu_time": 1.8385654562499100e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.9773882733678255e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"family_index": 3,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 8,
"real_time": 9.5071386545896530e+07,
"cpu_time": 1.8028767475000507e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.6470293692500315e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.0015753991901875e+08,
"cpu_time": 1.8153221362500036e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.3808856548034830e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time_median",
"family_index": 3,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 9.5071386545896530e+07,
"cpu_time": 1.8028767475000507e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.6470293692500315e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.0202777180208210e+06,
"cpu_time": 1.9240621422039159e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.6774867878493148e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 7.0092353742883320e-02,
"cpu_time": 1.0599012174106696e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.8343522307827628e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"family_index": 3,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 4,
"real_time": 1.7986493557691574e+08,
"cpu_time": 3.1871025724998957e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.9697117759833469e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"family_index": 3,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 4,
"real_time": 1.8023148924112320e+08,
"cpu_time": 3.2264847699997288e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.9575706138869638e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"family_index": 3,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 4,
"real_time": 1.7977836728096008e+08,
"cpu_time": 3.2172579575001234e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.9725863586353617e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"family_index": 3,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 4,
"real_time": 1.8014448881149292e+08,
"cpu_time": 3.2239760400000250e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.9604478109990177e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"family_index": 3,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 4,
"real_time": 1.8008193373680115e+08,
"cpu_time": 3.2149539849999088e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.9625182921976395e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time_mean",
"family_index": 3,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8002024292945862e+08,
"cpu_time": 3.2139550649999368e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.9645669703404665e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time_median",
"family_index": 3,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8008193373680115e+08,
"cpu_time": 3.2172579575001234e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.9625182921976395e+09,
"dst_gpu": 3.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time_stddev",
"family_index": 3,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9137136816232878e+05,
"cpu_time": 1.5735125840236386e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.3421836366727827e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time_cv",
"family_index": 3,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/3/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.0630547156706045e-03,
"cpu_time": 4.8958761158774001e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.0633099885054625e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"family_index": 4,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 381,
"real_time": 1.8480922752758183e+06,
"cpu_time": 2.0696343543304799e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2163395490567121e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"family_index": 4,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 381,
"real_time": 1.8468097289929441e+06,
"cpu_time": 2.0657908897638493e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2178787211791049e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"family_index": 4,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 381,
"real_time": 1.8496372757386051e+06,
"cpu_time": 2.0681680813648563e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2144882424930413e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"family_index": 4,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 381,
"real_time": 1.8469693242586611e+06,
"cpu_time": 2.0655158608922868e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2176870759042292e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"family_index": 4,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 381,
"real_time": 1.8425821139140476e+06,
"cpu_time": 2.0657112414698419e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2229674157094685e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8468181436360155e+06,
"cpu_time": 2.0669640855642627e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2178722008685111e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time_median",
"family_index": 4,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8469693242586609e+06,
"cpu_time": 2.0657908897638493e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2176870759042292e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.6238231379684366e+03,
"cpu_time": 1.8454879152341225e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.1541926029591946e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.4207263162374255e-03,
"cpu_time": 8.9284953140843790e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4221705839155311e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"family_index": 4,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 382,
"real_time": 1.8437452347001673e+06,
"cpu_time": 2.0674568848169118e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4431301276459685e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"family_index": 4,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 382,
"real_time": 1.8488204246912582e+06,
"cpu_time": 2.0691145000000093e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4309333078511469e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"family_index": 4,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 382,
"real_time": 1.8504321724417161e+06,
"cpu_time": 2.0700248455498491e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4270739138686405e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"family_index": 4,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 382,
"real_time": 1.8471266099672972e+06,
"cpu_time": 2.0664947905759318e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4349964727891805e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"family_index": 4,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 382,
"real_time": 1.8444091102723349e+06,
"cpu_time": 2.0682013455495902e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4415308699003421e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8469067104145549e+06,
"cpu_time": 2.0682584732984588e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4355329384110561e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time_median",
"family_index": 4,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8471266099672974e+06,
"cpu_time": 2.0682013455495904e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4349964727891805e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.8448407606427027e+03,
"cpu_time": 1.3792422794761083e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.8317633019926216e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.5403272642851315e-03,
"cpu_time": 6.6686166032067199e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.5402350510872248e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"family_index": 4,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 381,
"real_time": 1.8451983781369205e+06,
"cpu_time": 2.0705319133859547e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8792620859242082e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"family_index": 4,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 381,
"real_time": 1.8482409171889026e+06,
"cpu_time": 2.0697228871392573e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8646452135251835e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"family_index": 4,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 381,
"real_time": 1.8519730093428167e+06,
"cpu_time": 2.0732410419947365e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8467811989408843e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"family_index": 4,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 381,
"real_time": 1.8498972075101566e+06,
"cpu_time": 2.0716325065616197e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8567083260003496e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"family_index": 4,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 381,
"real_time": 1.8461835754919478e+06,
"cpu_time": 2.0716400577428341e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8745237567364872e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8482986175341487e+06,
"cpu_time": 2.0713536813648802e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8643841162254233e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time_median",
"family_index": 4,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8482409171889026e+06,
"cpu_time": 2.0716325065616197e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8646452135251835e+06,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.7452168204448599e+03,
"cpu_time": 1.3282203774022912e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.3162588236438578e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.4852669338179278e-03,
"cpu_time": 6.4123302039228991e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4848846872898586e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"family_index": 4,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 380,
"real_time": 1.8542003011756795e+06,
"cpu_time": 2.0823200526314466e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7672308638512801e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"family_index": 4,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 380,
"real_time": 1.8534533250395295e+06,
"cpu_time": 2.0768617552631130e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7679430907331396e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"family_index": 4,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 380,
"real_time": 1.8550386246620053e+06,
"cpu_time": 2.0783741473684660e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7664322221846163e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"family_index": 4,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 380,
"real_time": 1.8526423931976559e+06,
"cpu_time": 2.0765228578948395e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7687169483066030e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"family_index": 4,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 380,
"real_time": 1.8486347998412766e+06,
"cpu_time": 2.0763791131580428e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7725512904340789e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8527938887832295e+06,
"cpu_time": 2.0780915852631810e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7685748831019435e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time_median",
"family_index": 4,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8534533250395295e+06,
"cpu_time": 2.0768617552631130e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7679430907331396e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.4886071857148677e+03,
"cpu_time": 2.4932772472075467e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.3784734045049780e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.3431646125242733e-03,
"cpu_time": 1.1997918017130048e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.3448530945623968e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"family_index": 4,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 380,
"real_time": 1.8565472606018088e+06,
"cpu_time": 2.0901602921053621e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5299936279971771e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"family_index": 4,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 380,
"real_time": 1.8607350978479523e+06,
"cpu_time": 2.0897776789474010e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5220488975457154e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"family_index": 4,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 380,
"real_time": 1.8586428605263601e+06,
"cpu_time": 2.0877947184209533e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5260135979776382e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"family_index": 4,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 380,
"real_time": 1.8557160926368516e+06,
"cpu_time": 2.0852103684208752e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5315746983083829e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"family_index": 4,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 380,
"real_time": 1.8563409477983920e+06,
"cpu_time": 2.0896819736842525e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5303859497214273e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8575964518822730e+06,
"cpu_time": 2.0885250063157689e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5280033543100677e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time_median",
"family_index": 4,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8565472606018085e+06,
"cpu_time": 2.0896819736842525e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5299936279971771e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0711659329261488e+03,
"cpu_time": 2.0677322486277044e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.9308756606352996e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.1149708704640715e-03,
"cpu_time": 9.9004428597925030e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.1141927220202480e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"family_index": 4,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 379,
"real_time": 1.8557904647631906e+06,
"cpu_time": 2.1024784696570435e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0628663358675867e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"family_index": 4,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 379,
"real_time": 1.8612747898065462e+06,
"cpu_time": 2.1041797044856166e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0420553009060591e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"family_index": 4,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 379,
"real_time": 1.8602599161243960e+06,
"cpu_time": 2.1050623113455730e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0458971278094873e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"family_index": 4,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 379,
"real_time": 1.8628431794876822e+06,
"cpu_time": 2.1055027176781143e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0361263601398438e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"family_index": 4,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 379,
"real_time": 1.8582601454366972e+06,
"cpu_time": 2.1042697941953344e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0534795852922767e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8596856991237025e+06,
"cpu_time": 2.1042985994723365e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0480849420030519e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time_median",
"family_index": 4,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8602599161243960e+06,
"cpu_time": 2.1042697941953344e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0458971278094873e+07,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.7399667265774442e+03,
"cpu_time": 1.1576911052460978e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.0389207905923338e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.4733493556833483e-03,
"cpu_time": 5.5015533704978682e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4740469207470627e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"family_index": 4,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 377,
"real_time": 1.8685159315287187e+06,
"cpu_time": 2.1327632148542753e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4029529830421513e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"family_index": 4,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 377,
"real_time": 1.8715045968532581e+06,
"cpu_time": 2.1331306153846267e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4007125627196860e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"family_index": 4,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 377,
"real_time": 1.8716743537064255e+06,
"cpu_time": 2.1334884482758809e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4005855210917616e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"family_index": 4,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 377,
"real_time": 1.8703638138564487e+06,
"cpu_time": 2.1325571034484990e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4015668933387506e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"family_index": 4,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 377,
"real_time": 1.8701507545178831e+06,
"cpu_time": 2.1318157082228134e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4017265686561167e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8704418900925468e+06,
"cpu_time": 2.1327510180372195e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4015089057696933e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time_median",
"family_index": 4,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8703638138564485e+06,
"cpu_time": 2.1327632148542753e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4015668933387506e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.2696933122825419e+03,
"cpu_time": 6.3254517113249381e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 9.5169590416266903e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 6.7881997243962459e-04,
"cpu_time": 2.9658650530835429e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.7905091451417364e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"family_index": 4,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 372,
"real_time": 1.8956940757268439e+06,
"cpu_time": 2.1964512634409764e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7656783165236109e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"family_index": 4,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 372,
"real_time": 1.8968609591504128e+06,
"cpu_time": 2.1973949677419933e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7639769666345179e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"family_index": 4,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 372,
"real_time": 1.8981048476322736e+06,
"cpu_time": 2.1993671075268053e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7621656446112829e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"family_index": 4,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 372,
"real_time": 1.8915448050510378e+06,
"cpu_time": 2.1939706397851082e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7717450762994409e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"family_index": 4,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 372,
"real_time": 1.8917659020786155e+06,
"cpu_time": 2.1941193333333638e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7714211331535691e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8947941179278367e+06,
"cpu_time": 2.1962606623656498e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7669974274444842e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time_median",
"family_index": 4,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8956940757268439e+06,
"cpu_time": 2.1964512634409764e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7656783165236109e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.9904353051137341e+03,
"cpu_time": 2.2804863754846997e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.3680288554449822e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.5782375915247722e-03,
"cpu_time": 1.0383495978242986e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.5786168834566509e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"family_index": 4,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 369,
"real_time": 1.9034304827083217e+06,
"cpu_time": 2.2815761165313064e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5088746845538568e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"family_index": 4,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 369,
"real_time": 1.9043441312917606e+06,
"cpu_time": 2.2805841788617615e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5062316876977837e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"family_index": 4,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 369,
"real_time": 1.9019887822700366e+06,
"cpu_time": 2.2801785907861548e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5130503911201692e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"family_index": 4,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 369,
"real_time": 1.9046450360314695e+06,
"cpu_time": 2.2806039186989833e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5053617874374104e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"family_index": 4,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 369,
"real_time": 1.9024787366587096e+06,
"cpu_time": 2.2807110542005086e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5116305890577042e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9033774337920598e+06,
"cpu_time": 2.2807307718157438e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5090298279733849e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time_median",
"family_index": 4,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9034304827083214e+06,
"cpu_time": 2.2806039186989833e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5088746845538568e+08,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.1489281162085083e+03,
"cpu_time": 5.1416961215488732e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.3255475519078056e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 6.0362600491670345e-04,
"cpu_time": 2.2544073088712039e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.0365393830717014e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"family_index": 4,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 358,
"real_time": 1.9457454382774923e+06,
"cpu_time": 2.4736868463687715e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0778141676418591e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"family_index": 4,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 358,
"real_time": 1.9468292343565498e+06,
"cpu_time": 2.4732928072627126e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0772141505739889e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"family_index": 4,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 358,
"real_time": 1.9515229155915640e+06,
"cpu_time": 2.4783676117317048e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0746233022656009e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"family_index": 4,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 358,
"real_time": 1.9471979506220149e+06,
"cpu_time": 2.4744999329608334e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0770101721450989e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"family_index": 4,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 358,
"real_time": 1.9474227856792635e+06,
"cpu_time": 2.4752015027933489e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0768858285020583e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9477436649053772e+06,
"cpu_time": 2.4750097402234743e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0767095242257211e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time_median",
"family_index": 4,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9471979506220147e+06,
"cpu_time": 2.4744999329608330e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0770101721450989e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.2086532201324444e+03,
"cpu_time": 2.0167972338433769e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2194957221409185e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.1339547702955784e-03,
"cpu_time": 8.1486436237672173e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.1326134808901937e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"family_index": 4,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 346,
"real_time": 2.0338412307975329e+06,
"cpu_time": 2.8966130664740754e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0622573367515428e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"family_index": 4,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 346,
"real_time": 2.0319137717543351e+06,
"cpu_time": 2.8925630346821821e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0642135794860415e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"family_index": 4,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 346,
"real_time": 2.0289297358299782e+06,
"cpu_time": 2.8920062687861924e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0672495089063435e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"family_index": 4,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 346,
"real_time": 2.0272788637777993e+06,
"cpu_time": 2.8918098612719094e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0689329302156222e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"family_index": 4,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 346,
"real_time": 2.0271993324300544e+06,
"cpu_time": 2.8933491098265708e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0690140988613012e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0298325869179401e+06,
"cpu_time": 2.8932682682081861e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0663334908441701e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time_median",
"family_index": 4,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0289297358299785e+06,
"cpu_time": 2.8925630346821821e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0672495089063435e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.9441139258828057e+03,
"cpu_time": 1.9628524180991913e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.9954502930678050e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.4504220421217562e-03,
"cpu_time": 6.7842046991197066e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4496451353765059e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"family_index": 4,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 325,
"real_time": 2.1614762621286972e+06,
"cpu_time": 3.7429450830771974e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.8809623528960743e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"family_index": 4,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 325,
"real_time": 2.1577642213266632e+06,
"cpu_time": 3.7390170307690613e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.8876388425990367e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"family_index": 4,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 325,
"real_time": 2.1456681593106342e+06,
"cpu_time": 3.7325332369230702e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9095551488703237e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"family_index": 4,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 325,
"real_time": 2.1565122726874859e+06,
"cpu_time": 3.7372111815385809e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.8898957850798416e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"family_index": 4,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 325,
"real_time": 2.1481006574602085e+06,
"cpu_time": 3.7329256369230277e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.9051279887033844e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.1539043145827381e+06,
"cpu_time": 3.7369264338461882e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.8946360236297321e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time_median",
"family_index": 4,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.1565122726874859e+06,
"cpu_time": 3.7372111815385809e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 3.8898957850798416e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 6.7184708983151386e+03,
"cpu_time": 4.3583933542926043e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2155876194960197e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 3.1192058313958414e-03,
"cpu_time": 1.1663042961771067e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.1211841417805035e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"family_index": 4,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 174,
"real_time": 4.0116933731083898e+06,
"cpu_time": 7.1560809137932230e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.1820783493730655e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"family_index": 4,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 174,
"real_time": 4.0215958610990611e+06,
"cpu_time": 7.1681443563217996e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.1717807008621097e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"family_index": 4,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 174,
"real_time": 4.0156206988526145e+06,
"cpu_time": 7.1570842873561354e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.1779882260278625e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"family_index": 4,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 174,
"real_time": 4.0199370243875631e+06,
"cpu_time": 7.1682940574718509e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.1735021962330384e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"family_index": 4,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 174,
"real_time": 4.0263931585432985e+06,
"cpu_time": 7.1699990402300674e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.1668101795776443e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.0190480231981860e+06,
"cpu_time": 7.1639205310346158e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.1744319304147449e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time_median",
"family_index": 4,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.0199370243875631e+06,
"cpu_time": 7.1681443563217996e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.1735021962330384e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.6362719156484700e+03,
"cpu_time": 6.7474167970603494e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.8544741642747046e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.4023897905960745e-03,
"cpu_time": 9.4186092207891711e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4024600860345186e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"family_index": 4,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 90,
"real_time": 7.8837633236414855e+06,
"cpu_time": 1.4056749711110367e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.2561440041431017e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"family_index": 4,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 90,
"real_time": 7.8676477747244965e+06,
"cpu_time": 1.4069561077777356e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.2648619969740553e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"family_index": 4,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 90,
"real_time": 7.9819861592517961e+06,
"cpu_time": 1.4172058533333458e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.2037697548632779e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"family_index": 4,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 90,
"real_time": 7.8680405496723121e+06,
"cpu_time": 1.4054622166666932e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.2646490937819929e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"family_index": 4,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 90,
"real_time": 7.8749028862350518e+06,
"cpu_time": 1.4061762988887949e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.2609327993938212e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.8952681387050273e+06,
"cpu_time": 1.4082950895555213e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.2500715298312502e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time_median",
"family_index": 4,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.8749028862350509e+06,
"cpu_time": 1.4061762988887947e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.2609327993938212e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.8916626713908445e+04,
"cpu_time": 5.0143306298083131e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.6124508488133512e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 6.1956890956121075e-03,
"cpu_time": 3.5605681415752924e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.1468397189942802e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"family_index": 4,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 46,
"real_time": 1.5262760383927304e+07,
"cpu_time": 2.7323593999999896e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.3969021534708805e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"family_index": 4,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 46,
"real_time": 1.5263143197997756e+07,
"cpu_time": 2.7314231000001084e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.3967918750053692e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"family_index": 4,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 46,
"real_time": 1.5292416245717069e+07,
"cpu_time": 2.7356930934783831e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.3883754484380522e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"family_index": 4,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 46,
"real_time": 1.5032098223657712e+07,
"cpu_time": 2.6056694934783556e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.4643710413216429e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"family_index": 4,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 46,
"real_time": 1.5294343192616235e+07,
"cpu_time": 2.7352231913045235e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.3878225533999166e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.5228952248783220e+07,
"cpu_time": 2.7080736556522720e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.4068526143271732e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time_median",
"family_index": 4,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.5263143197997754e+07,
"cpu_time": 2.7323593999999892e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.3967918750053692e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.1109362342198717e+05,
"cpu_time": 5.7274638786231936e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.2450533389386375e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 7.2948960379637072e-03,
"cpu_time": 2.1149586779771929e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.3636529807886118e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"family_index": 4,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 23,
"real_time": 3.0112891984374627e+07,
"cpu_time": 5.2410173739130899e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.4571517099601278e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"family_index": 4,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 23,
"real_time": 3.0117247334640957e+07,
"cpu_time": 5.2338769260868303e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.4565071471728535e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"family_index": 4,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 23,
"real_time": 3.0122561623220857e+07,
"cpu_time": 5.2422950739133224e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.4557209203793058e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"family_index": 4,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 23,
"real_time": 3.0092251689537711e+07,
"cpu_time": 5.2379639521736071e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.4602088731918983e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"family_index": 4,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 23,
"real_time": 3.0144454668397490e+07,
"cpu_time": 5.2405086347829163e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.4524848592039614e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.0117881460034333e+07,
"cpu_time": 5.2391323921739534e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.4564147019816294e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time_median",
"family_index": 4,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.0117247334640957e+07,
"cpu_time": 5.2405086347829171e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.4565071471728535e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8776457646973373e+04,
"cpu_time": 3.3329765778772940e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.7781495290498673e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 6.2343221822853826e-04,
"cpu_time": 6.3616956556699863e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.2340462341049872e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"family_index": 4,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 10,
"real_time": 5.9363515675067902e+07,
"cpu_time": 9.9733661099992335e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.5218928317741175e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"family_index": 4,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 10,
"real_time": 5.9164601191878319e+07,
"cpu_time": 9.9633707600003153e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.5370956719446096e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"family_index": 4,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 10,
"real_time": 5.9197128564119339e+07,
"cpu_time": 9.9546580099990934e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.5346026489991703e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"family_index": 4,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 10,
"real_time": 5.9718941897153854e+07,
"cpu_time": 9.8046241800000191e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.4949801096993876e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"family_index": 4,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 10,
"real_time": 5.9126920625567436e+07,
"cpu_time": 9.9608335999994323e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.5399870847311497e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.9314221590757385e+07,
"cpu_time": 9.9313705319996193e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.5257116694296875e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time_median",
"family_index": 4,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.9197128564119339e+07,
"cpu_time": 9.9608335999994323e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.5346026489991703e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.4362221027860470e+05,
"cpu_time": 7.1173485084477556e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.8516511533361785e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 4.1073153072713857e-03,
"cpu_time": 7.1665320365554043e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.0914032722051789e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"family_index": 4,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 8,
"real_time": 1.1534585431218147e+08,
"cpu_time": 1.8806281975000161e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.6544448016915159e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"family_index": 4,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 8,
"real_time": 1.1593741737306118e+08,
"cpu_time": 1.9259290425000587e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.6306958026541786e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"family_index": 4,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 8,
"real_time": 1.1553475540131330e+08,
"cpu_time": 1.8814828637499657e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.6468347133740263e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"family_index": 4,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 8,
"real_time": 1.1599481571465731e+08,
"cpu_time": 1.9273312612499183e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.6284043704218760e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"family_index": 4,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 8,
"real_time": 1.1540485359728336e+08,
"cpu_time": 1.8758924987500337e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.6520652751180124e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.1564353927969933e+08,
"cpu_time": 1.8982527727499989e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.6424889926519222e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time_median",
"family_index": 4,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.1553475540131330e+08,
"cpu_time": 1.8814828637499660e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.6468347133740263e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.0297714211793581e+05,
"cpu_time": 2.5996979414341855e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2155119887691770e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.6199227730755036e-03,
"cpu_time": 1.3695214771992686e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.6182334318790531e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"family_index": 4,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 4,
"real_time": 2.0591312646865845e+08,
"cpu_time": 3.2382666050000352e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.2145380064608545e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"family_index": 4,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 4,
"real_time": 2.0644620433449745e+08,
"cpu_time": 3.2431641425000632e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.2010732164407063e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"family_index": 4,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 4,
"real_time": 2.0662897452712059e+08,
"cpu_time": 3.2549405199998206e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.1964726943900528e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"family_index": 4,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 4,
"real_time": 2.0605592802166939e+08,
"cpu_time": 3.2415575824998653e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.2109242102808247e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"family_index": 4,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 4,
"real_time": 2.0650283992290497e+08,
"cpu_time": 3.2501045950002092e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.1996467670898228e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time_mean",
"family_index": 4,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0630941465497017e+08,
"cpu_time": 3.2456066889999992e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.2045309789324532e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time_median",
"family_index": 4,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0644620433449745e+08,
"cpu_time": 3.2431641425000632e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.2010732164407063e+09,
"dst_gpu": 4.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time_stddev",
"family_index": 4,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.0803533176601405e+05,
"cpu_time": 6.7747636396501435e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.7743424728886234e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time_cv",
"family_index": 4,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/4/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.4930745273121406e-03,
"cpu_time": 2.0873643324100708e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4937642804622687e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"family_index": 5,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 382,
"real_time": 1.8465842129072938e+06,
"cpu_time": 2.0683208115184263e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2181495820064377e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"family_index": 5,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 382,
"real_time": 1.8511216136718630e+06,
"cpu_time": 2.0724831780104109e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2127125358745190e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"family_index": 5,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 382,
"real_time": 1.8467932190943854e+06,
"cpu_time": 2.0680959921466508e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2178985484950836e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"family_index": 5,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 382,
"real_time": 1.8412949712143990e+06,
"cpu_time": 2.0684308089007146e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2245213635154520e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"family_index": 5,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 382,
"real_time": 1.8414499327501997e+06,
"cpu_time": 2.0690049214659233e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2243341657856731e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8454487899276279e+06,
"cpu_time": 2.0692671424084255e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2195232391354330e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time_median",
"family_index": 5,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8465842129072933e+06,
"cpu_time": 2.0684308089007144e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2181495820064377e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.1389211453164153e+03,
"cpu_time": 1.8288133882577890e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.9759043158994054e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.2427721472990476e-03,
"cpu_time": 8.8379762611473576e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.2418798002032457e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"family_index": 5,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 382,
"real_time": 1.8465305973089384e+06,
"cpu_time": 2.0693080811518996e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4364279757609768e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"family_index": 5,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 382,
"real_time": 1.8515576493930309e+06,
"cpu_time": 2.0737806910992803e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4243828987368904e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"family_index": 5,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 382,
"real_time": 1.8475077786825411e+06,
"cpu_time": 2.0696796701570686e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4340814661368951e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"family_index": 5,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 382,
"real_time": 1.8426901457056585e+06,
"cpu_time": 2.0699525366491647e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4456741786410715e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"family_index": 5,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 382,
"real_time": 1.8423789499319312e+06,
"cpu_time": 2.0701090628271522e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4464250963693773e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8461330242044199e+06,
"cpu_time": 2.0705660083769131e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4373983231290430e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time_median",
"family_index": 5,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8465305973089382e+06,
"cpu_time": 2.0699525366491645e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4364279757609768e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.7888342100649861e+03,
"cpu_time": 1.8225880050113374e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 9.1010549659993067e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.0523083441929984e-03,
"cpu_time": 8.8023661049088588e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.0509889586792095e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"family_index": 5,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 381,
"real_time": 1.8483568434747122e+06,
"cpu_time": 2.0725939737532800e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8640892357126456e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"family_index": 5,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 381,
"real_time": 1.8535680207469699e+06,
"cpu_time": 2.0782530971129788e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8391684667700548e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"family_index": 5,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 381,
"real_time": 1.8483190651750192e+06,
"cpu_time": 2.0722750787401879e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8642704112607222e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"family_index": 5,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 381,
"real_time": 1.8445302544679893e+06,
"cpu_time": 2.0731965459317460e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8824783222249579e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"family_index": 5,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 381,
"real_time": 1.8483438246476762e+06,
"cpu_time": 2.0725053018374669e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8641516700081769e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8486236017024736e+06,
"cpu_time": 2.0737647994751320e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8628316211953126e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time_median",
"family_index": 5,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8483438246476762e+06,
"cpu_time": 2.0725939737532798e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8641516700081769e+06,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.2188875762509256e+03,
"cpu_time": 2.5320032983780834e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.5421834996148724e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.7412347074258489e-03,
"cpu_time": 1.2209693688594440e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.7400573152340686e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"family_index": 5,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 377,
"real_time": 1.8511144930889907e+06,
"cpu_time": 2.0782076870025520e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7701768379177563e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"family_index": 5,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 377,
"real_time": 1.8551378677718327e+06,
"cpu_time": 2.0826713660475174e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7663377245032985e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"family_index": 5,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 377,
"real_time": 1.8495802567978092e+06,
"cpu_time": 2.0786585039790205e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7716452086665034e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"family_index": 5,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 377,
"real_time": 1.8444462418665835e+06,
"cpu_time": 2.0785867267904447e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7765765819685109e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"family_index": 5,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 377,
"real_time": 1.8488328803709859e+06,
"cpu_time": 2.0779301644562962e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7723613825725984e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8498223479792406e+06,
"cpu_time": 2.0792108896551661e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7714195471257340e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time_median",
"family_index": 5,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8495802567978096e+06,
"cpu_time": 2.0785867267904449e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7716452086665034e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.8686095975724661e+03,
"cpu_time": 1.9568008762149177e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.7048020474699457e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.0913411505697095e-03,
"cpu_time": 9.4112669664761592e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.0914311651812102e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"family_index": 5,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 376,
"real_time": 1.8501916252678209e+06,
"cpu_time": 2.0847361329786996e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5421195893973127e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"family_index": 5,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 376,
"real_time": 1.8559700306371830e+06,
"cpu_time": 2.0913366542555543e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5310915003029704e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"family_index": 5,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 376,
"real_time": 1.8500712571465634e+06,
"cpu_time": 2.0849889707448562e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5423500444560558e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"family_index": 5,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 376,
"real_time": 1.8460308154781740e+06,
"cpu_time": 2.0857780558511431e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5501032512842603e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"family_index": 5,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 376,
"real_time": 1.8501065833527506e+06,
"cpu_time": 2.0850115664891389e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5422824063052684e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8504740623764987e+06,
"cpu_time": 2.0863702760638786e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5415893583491735e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time_median",
"family_index": 5,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8501065833527506e+06,
"cpu_time": 2.0850115664891389e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5422824063052684e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.5470119017843012e+03,
"cpu_time": 2.8035761464586894e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.7829105155898971e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.9168125476068541e-03,
"cpu_time": 1.3437577109983961e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.9152165396023179e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"family_index": 5,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 376,
"real_time": 1.8505312777864325e+06,
"cpu_time": 2.0996436223402298e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0829389145362422e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"family_index": 5,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 376,
"real_time": 1.8549917113751746e+06,
"cpu_time": 2.1040223723406252e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0659075831035078e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"family_index": 5,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 376,
"real_time": 1.8470027092955709e+06,
"cpu_time": 2.0994837473404850e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0964703700943455e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"family_index": 5,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 376,
"real_time": 1.8501074259705655e+06,
"cpu_time": 2.0996872446810333e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0845615859976187e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"family_index": 5,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 376,
"real_time": 1.8469418861214046e+06,
"cpu_time": 2.0998097047873535e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0967040698423073e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8499150021098300e+06,
"cpu_time": 2.1005293382979454e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0853165047148049e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time_median",
"family_index": 5,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8501074259705651e+06,
"cpu_time": 2.0996872446810333e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0845615859976187e+07,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.2981308239959458e+03,
"cpu_time": 1.9561463443958007e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2619572376372346e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.7828553313176143e-03,
"cpu_time": 9.3126351949973814e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.7810880245046025e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"family_index": 5,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 380,
"real_time": 1.8593009536025897e+06,
"cpu_time": 2.1291140947366063e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4099062311137345e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"family_index": 5,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 380,
"real_time": 1.8576057919601321e+06,
"cpu_time": 2.1273989052630034e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4111928436839527e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"family_index": 5,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 380,
"real_time": 1.8590386645212821e+06,
"cpu_time": 2.1285741105262595e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4101051527484196e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"family_index": 5,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 380,
"real_time": 1.8578790737879709e+06,
"cpu_time": 2.1279806026316956e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4109852664712074e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"family_index": 5,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 380,
"real_time": 1.8574335843272579e+06,
"cpu_time": 2.1278603473684886e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4113236791448760e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8582516136398464e+06,
"cpu_time": 2.1281856121052108e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4107026346324381e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time_median",
"family_index": 5,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8578790737879705e+06,
"cpu_time": 2.1279806026316956e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4109852664712074e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 8.5813853509059720e+02,
"cpu_time": 6.6703536399066138e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.5136877458164970e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 4.6179889138356237e-04,
"cpu_time": 3.1342912958180703e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.6173357771559371e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"family_index": 5,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 374,
"real_time": 1.8801808518788374e+06,
"cpu_time": 2.1903311711229249e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7884977100797862e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"family_index": 5,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 374,
"real_time": 1.8738428498668110e+06,
"cpu_time": 2.1858072807487543e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7979293996679890e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"family_index": 5,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 374,
"real_time": 1.8788503462238295e+06,
"cpu_time": 2.1890416363636605e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7904723814418209e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"family_index": 5,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 374,
"real_time": 1.8816713637013715e+06,
"cpu_time": 2.1915268663102123e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7862888818624043e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"family_index": 5,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 374,
"real_time": 1.8834852625025725e+06,
"cpu_time": 2.1915159893049905e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7836055340479940e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8796061348346844e+06,
"cpu_time": 2.1896445887701083e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7893587814199990e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time_median",
"family_index": 5,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8801808518788372e+06,
"cpu_time": 2.1903311711229249e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7884977100797862e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6547835592560705e+03,
"cpu_time": 2.3772351842871863e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.4304968608774652e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.9444411738832276e-03,
"cpu_time": 1.0856717096825493e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.9468620878211023e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"family_index": 5,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 376,
"real_time": 1.8741139823024685e+06,
"cpu_time": 2.2706852765956097e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5950492334076583e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"family_index": 5,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 376,
"real_time": 1.8777918725236112e+06,
"cpu_time": 2.2731451010637912e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5840906297607553e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"family_index": 5,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 376,
"real_time": 1.8734845989650025e+06,
"cpu_time": 2.2699508111703559e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5969288489442647e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"family_index": 5,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 376,
"real_time": 1.8795914437121495e+06,
"cpu_time": 2.2739890797871058e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5787442718354094e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"family_index": 5,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 376,
"real_time": 1.8741803730161802e+06,
"cpu_time": 2.2707517819147245e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5948510351353860e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8758324541038827e+06,
"cpu_time": 2.2717044101063171e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5899328038166940e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time_median",
"family_index": 5,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8741803730161800e+06,
"cpu_time": 2.2707517819147245e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5948510351353860e+08,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.7001974196628680e+03,
"cpu_time": 1.7547470473341248e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.0409225858728425e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.4394662027280036e-03,
"cpu_time": 7.7243634318251880e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4384649812574960e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"family_index": 5,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 367,
"real_time": 1.9180650874523178e+06,
"cpu_time": 2.4724894087191513e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0933685273347816e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"family_index": 5,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 367,
"real_time": 1.9170274531052574e+06,
"cpu_time": 2.4719248583105160e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0939603377108512e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"family_index": 5,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 367,
"real_time": 1.9192116511719204e+06,
"cpu_time": 2.4721809782016124e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0927153337775042e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"family_index": 5,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 367,
"real_time": 1.9185245826926681e+06,
"cpu_time": 2.4719481198911672e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0931066606697457e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"family_index": 5,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 367,
"real_time": 1.9160635699264260e+06,
"cpu_time": 2.4722161144412332e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0945106586836927e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9177784688697183e+06,
"cpu_time": 2.4721518959127362e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0935323036353152e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time_median",
"family_index": 5,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9180650874523178e+06,
"cpu_time": 2.4721809782016124e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0933685273347816e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.2447640960998162e+03,
"cpu_time": 2.3019308564146294e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.0989895912023983e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 6.4906563312990126e-04,
"cpu_time": 9.3114458711880251e-05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.4917968747723969e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"family_index": 5,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 349,
"real_time": 2.0130427786413271e+06,
"cpu_time": 2.8910676361029074e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0835642662451925e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"family_index": 5,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 349,
"real_time": 2.0134091186457258e+06,
"cpu_time": 2.8902243925501243e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0831851615041876e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"family_index": 5,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 349,
"real_time": 2.0105927858373583e+06,
"cpu_time": 2.8899593667621571e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0861031779009316e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"family_index": 5,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 349,
"real_time": 2.0120204469712081e+06,
"cpu_time": 2.8910821948426035e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0846229501861618e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"family_index": 5,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 349,
"real_time": 2.0097575225816627e+06,
"cpu_time": 2.8902853123209807e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0869701707159910e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0117645305354570e+06,
"cpu_time": 2.8905237805157551e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0848891453104928e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time_median",
"family_index": 5,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0120204469712086e+06,
"cpu_time": 2.8902853123209807e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0846229501861618e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.5656586519223940e+03,
"cpu_time": 5.1784857360161959e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.6228267827220501e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 7.7825144451954020e-04,
"cpu_time": 1.7915388798815562e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.7837557280791068e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"family_index": 5,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 342,
"real_time": 2.0459876392358011e+06,
"cpu_time": 3.6549505292396401e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.1000286801015267e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"family_index": 5,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 342,
"real_time": 2.0413073470777893e+06,
"cpu_time": 3.6527593128655278e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.1094291910567112e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"family_index": 5,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 342,
"real_time": 2.0431291631270438e+06,
"cpu_time": 3.6574717807016037e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.1057648979769311e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"family_index": 5,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 342,
"real_time": 2.0479277700147650e+06,
"cpu_time": 3.6541516345030349e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0961444650655427e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"family_index": 5,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 342,
"real_time": 2.0484733014035295e+06,
"cpu_time": 3.6551721783626280e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0950536159062810e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0453650441717855e+06,
"cpu_time": 3.6549010871344870e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.1012841700213985e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time_median",
"family_index": 5,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0459876392358008e+06,
"cpu_time": 3.6549505292396396e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.1000286801015267e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.0854708364498656e+03,
"cpu_time": 1.7199765002632817e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.1893238167153606e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.5085184159384331e-03,
"cpu_time": 4.7059454120871328e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.5091185004825129e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"family_index": 5,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 189,
"real_time": 3.7035002107066768e+06,
"cpu_time": 6.8695707460319046e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5300972176261024e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"family_index": 5,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 189,
"real_time": 3.7059103170774444e+06,
"cpu_time": 6.8878807671958692e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5271511085111341e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"family_index": 5,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 189,
"real_time": 3.7058799012647890e+06,
"cpu_time": 6.8893471164021138e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5271882648636456e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"family_index": 5,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 189,
"real_time": 3.7128911193499649e+06,
"cpu_time": 6.8993391481482163e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5186393731193695e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"family_index": 5,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 189,
"real_time": 3.7027890769341006e+06,
"cpu_time": 6.8877987619047156e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5309672388607912e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.7061941250665956e+06,
"cpu_time": 6.8867873079365641e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5268086405962095e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time_median",
"family_index": 5,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.7058799012647890e+06,
"cpu_time": 6.8878807671958702e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5271882648636456e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.9962652367073542e+03,
"cpu_time": 1.0755580041612175e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.8761411056203041e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.0782665726220013e-03,
"cpu_time": 1.5617703234739200e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.0771697000600594e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"family_index": 5,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 95,
"real_time": 7.4726483980683908e+06,
"cpu_time": 1.3944482599999661e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4902998525493994e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"family_index": 5,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 95,
"real_time": 7.4799326444534883e+06,
"cpu_time": 1.3947055947368139e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4859270256773300e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"family_index": 5,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 95,
"real_time": 7.7806822318387656e+06,
"cpu_time": 1.3941250221053354e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.3125308295838556e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"family_index": 5,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 95,
"real_time": 7.4766584966135649e+06,
"cpu_time": 1.3945731252631862e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4878914845713434e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"family_index": 5,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 95,
"real_time": 7.4719663807436042e+06,
"cpu_time": 1.3941099736842645e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4907097128374243e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.5363776303435620e+06,
"cpu_time": 1.3943923951579133e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4534717810438709e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time_median",
"family_index": 5,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.4766584966135649e+06,
"cpu_time": 1.3944482599999661e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.4878914845713434e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.3660835790565636e+05,
"cpu_time": 2.6698710934991873e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.8812088394919723e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.8126527704189467e-02,
"cpu_time": 1.9147200621363308e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.7696775071166293e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"family_index": 5,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 49,
"real_time": 1.4596219364629716e+07,
"cpu_time": 2.7057142326528411e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5976880946734419e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"family_index": 5,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 49,
"real_time": 1.4625724728162192e+07,
"cpu_time": 2.7100937122447841e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5884128990052872e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"family_index": 5,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 49,
"real_time": 1.4618100427395226e+07,
"cpu_time": 2.7086246714284986e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5908060580999861e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"family_index": 5,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 49,
"real_time": 1.4627439026929894e+07,
"cpu_time": 2.7120765367346060e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5878751486469374e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"family_index": 5,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 49,
"real_time": 1.4601695218256542e+07,
"cpu_time": 2.7084106571430068e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5959638930206947e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4613835753074717e+07,
"cpu_time": 2.7089839620407481e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5921492186892700e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time_median",
"family_index": 5,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4618100427395228e+07,
"cpu_time": 2.7086246714284986e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5908060580999861e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4162573588869040e+04,
"cpu_time": 2.3422458380691449e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.4514808567037554e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 9.6912089530561954e-04,
"cpu_time": 8.6462152264078752e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 9.6936764131857513e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"family_index": 5,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 25,
"real_time": 2.8262508660554886e+07,
"cpu_time": 5.1237790079999283e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7489672488742504e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"family_index": 5,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 25,
"real_time": 2.5042227208614349e+07,
"cpu_time": 4.9233317079997502e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 5.3596561872032709e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"family_index": 5,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 25,
"real_time": 2.8291513621807098e+07,
"cpu_time": 5.1304395080001086e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7440985234718933e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"family_index": 5,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 25,
"real_time": 2.8223506882786751e+07,
"cpu_time": 5.1166709079998329e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7555297985261402e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"family_index": 5,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 25,
"real_time": 2.8248172327876091e+07,
"cpu_time": 5.1222821999999724e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7513774145150681e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.7613585740327835e+07,
"cpu_time": 5.0833006663999192e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.8719258345181246e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time_median",
"family_index": 5,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.8248172327876091e+07,
"cpu_time": 5.1222821999999724e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7513774145150681e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4376435410606305e+06,
"cpu_time": 8.9559710757649131e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.7268089909412926e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 5.2062906808985912e-02,
"cpu_time": 1.7618416976518697e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.5969837874410039e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"family_index": 5,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 10,
"real_time": 5.4906305670738220e+07,
"cpu_time": 9.5782241400002033e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8889731829664888e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"family_index": 5,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 10,
"real_time": 5.2032082155346870e+07,
"cpu_time": 9.2615387800003648e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.1590373646505184e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"family_index": 5,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 10,
"real_time": 5.4992497712373734e+07,
"cpu_time": 9.5815661000006005e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8813104908235502e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"family_index": 5,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 10,
"real_time": 5.5000162869691849e+07,
"cpu_time": 9.6020878100000575e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8806302016956911e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"family_index": 5,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 10,
"real_time": 5.4992417618632317e+07,
"cpu_time": 9.5991630200001046e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8813176002113018e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.4384693205356613e+07,
"cpu_time": 9.5245159700002670e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9382537680695105e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time_median",
"family_index": 5,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.4992417618632317e+07,
"cpu_time": 9.5815661000006005e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.8813176002113018e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.3157143027291538e+06,
"cpu_time": 1.4738203736602277e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2346934034696193e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.4192731910080225e-02,
"cpu_time": 1.5473966113368664e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.5002631728913611e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"family_index": 5,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 8,
"real_time": 1.0720609128475189e+08,
"cpu_time": 1.8399000837499103e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0078396252131624e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"family_index": 5,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 8,
"real_time": 1.0387394763529301e+08,
"cpu_time": 1.7877492612498713e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.1684847280954657e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"family_index": 5,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 8,
"real_time": 1.0741833131760359e+08,
"cpu_time": 1.8435018425000748e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 4.9979450007711878e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"family_index": 5,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 8,
"real_time": 1.0411288961768150e+08,
"cpu_time": 1.7962440137500834e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.1566229116440077e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"family_index": 5,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 8,
"real_time": 1.0735198855400085e+08,
"cpu_time": 1.8430539774999487e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0010336951507883e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.0599264968186617e+08,
"cpu_time": 1.8220898357499781e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0663851921749220e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time_median",
"family_index": 5,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.0720609128475189e+08,
"cpu_time": 1.8399000837499100e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0078396252131624e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8286065104648801e+06,
"cpu_time": 2.7669729751309012e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.7962524156262815e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.7252201128600791e-02,
"cpu_time": 1.5185711049159147e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.7361989035520185e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"family_index": 5,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 4,
"real_time": 2.0024105906486511e+08,
"cpu_time": 3.1605518474998462e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.3622460299322395e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"family_index": 5,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 4,
"real_time": 2.0016381517052650e+08,
"cpu_time": 3.1430412149998689e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.3643153388400497e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"family_index": 5,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 4,
"real_time": 2.0056834071874619e+08,
"cpu_time": 3.1479226499999416e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.3534960709760828e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"family_index": 5,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 4,
"real_time": 1.9983509555459023e+08,
"cpu_time": 3.1487899025000840e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.3731393928584433e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"family_index": 5,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 4,
"real_time": 2.0072638243436813e+08,
"cpu_time": 3.1603766949999112e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.3492810012210693e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time_mean",
"family_index": 5,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0030693858861923e+08,
"cpu_time": 3.1521364619999301e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.3604955667655773e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time_median",
"family_index": 5,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0024105906486511e+08,
"cpu_time": 3.1487899025000846e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.3622460299322395e+09,
"dst_gpu": 5.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time_stddev",
"family_index": 5,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.5061534712040267e+05,
"cpu_time": 7.9120057990826736e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 9.3845689820342846e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time_cv",
"family_index": 5,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/5/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.7503904237709887e-03,
"cpu_time": 2.5100454547144690e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.7506905593239316e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"family_index": 6,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 383,
"real_time": 1.8421500724551568e+06,
"cpu_time": 2.0726039660572810e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2234887706738170e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"family_index": 6,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 383,
"real_time": 1.8388882782861928e+06,
"cpu_time": 2.0693747728461192e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2274327637877981e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"family_index": 6,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 383,
"real_time": 1.8347040155281522e+06,
"cpu_time": 2.0696619712794907e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2325126916021346e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"family_index": 6,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 383,
"real_time": 1.8399075880093807e+06,
"cpu_time": 2.0691840757179037e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2261987649235763e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"family_index": 6,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 383,
"real_time": 1.8438532814259189e+06,
"cpu_time": 2.0737469738903223e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2214348838170106e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8399006471409600e+06,
"cpu_time": 2.0709143519582234e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2262135749608674e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time_median",
"family_index": 6,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8399075880093810e+06,
"cpu_time": 2.0696619712794907e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2261987649235763e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.4897403566607813e+03,
"cpu_time": 2.1101634940485119e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.2256939855393966e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.8967004343867823e-03,
"cpu_time": 1.0189525665574607e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.8981530043062811e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"family_index": 6,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 383,
"real_time": 1.8442301836862369e+06,
"cpu_time": 2.0753741253262507e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4419617857169416e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"family_index": 6,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 383,
"real_time": 1.8392634461899821e+06,
"cpu_time": 2.0701933315924972e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4539568363464484e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"family_index": 6,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 383,
"real_time": 1.8374140940443370e+06,
"cpu_time": 2.0737829425585889e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4584397314426647e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"family_index": 6,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 383,
"real_time": 1.8388476525874098e+06,
"cpu_time": 2.0702373315929098e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4549639490107745e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"family_index": 6,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 383,
"real_time": 1.8436713067814100e+06,
"cpu_time": 2.0768846605744101e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4433082892097440e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8406853366578750e+06,
"cpu_time": 2.0732944783289314e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4505261183453146e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time_median",
"family_index": 6,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8392634461899821e+06,
"cpu_time": 2.0737829425585889e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4539568363464484e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.0652015233736497e+03,
"cpu_time": 3.0172916140780453e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.4083447897903661e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.6652501447852700e-03,
"cpu_time": 1.4553126174869152e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.6645997782717776e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"family_index": 6,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 383,
"real_time": 1.8473630021408310e+06,
"cpu_time": 2.0796710652739564e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8688579239777308e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"family_index": 6,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 383,
"real_time": 1.8414918382266990e+06,
"cpu_time": 2.0746948851175616e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8971341929906663e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"family_index": 6,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 383,
"real_time": 1.8373890632683565e+06,
"cpu_time": 2.0754014334202693e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.9170009376544692e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"family_index": 6,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 383,
"real_time": 1.8424472651560183e+06,
"cpu_time": 2.0749037911228430e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8925204589845371e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"family_index": 6,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 383,
"real_time": 1.8571146959761113e+06,
"cpu_time": 2.0903057467363426e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8222876247223187e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8451611729536031e+06,
"cpu_time": 2.0789953843341947e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8795602276659440e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time_median",
"family_index": 6,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8424472651560181e+06,
"cpu_time": 2.0754014334202693e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.8925204589845371e+06,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.5658734831055363e+03,
"cpu_time": 6.6432814575144166e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.6308776382878495e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 4.1003862394278666e-03,
"cpu_time": 3.1954286707769433e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.0890286739372131e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"family_index": 6,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 382,
"real_time": 1.8490496928013095e+06,
"cpu_time": 2.0860477905760049e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7721535623175435e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"family_index": 6,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 382,
"real_time": 1.8426052712601018e+06,
"cpu_time": 2.0802116570678507e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7783515824629635e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"family_index": 6,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 382,
"real_time": 1.8387970938429565e+06,
"cpu_time": 2.0812491780103273e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7820345762847163e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"family_index": 6,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 382,
"real_time": 1.8426823443172411e+06,
"cpu_time": 2.0799701832460540e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7782772001400679e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"family_index": 6,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 382,
"real_time": 1.8475100154109562e+06,
"cpu_time": 2.0845806675391947e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7736304391676683e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8441288835265129e+06,
"cpu_time": 2.0824118952878867e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7768894720745917e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time_median",
"family_index": 6,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8426823443172411e+06,
"cpu_time": 2.0812491780103273e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7782772001400679e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.1378819457939671e+03,
"cpu_time": 2.7421809300212767e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.9870349265423734e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.2438138585417784e-03,
"cpu_time": 1.3168292671715551e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.2438283242723848e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"family_index": 6,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 382,
"real_time": 1.8490090714156546e+06,
"cpu_time": 2.0913382958114964e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5443849904870257e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"family_index": 6,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 382,
"real_time": 1.8442446537832706e+06,
"cpu_time": 2.0867850680626852e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5535415469720848e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"family_index": 6,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 382,
"real_time": 1.8490266731189911e+06,
"cpu_time": 2.0968170575916104e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5443512499174498e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"family_index": 6,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 382,
"real_time": 1.8467041306694085e+06,
"cpu_time": 2.0898198429320457e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5488088704411991e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"family_index": 6,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 382,
"real_time": 1.8481089780320467e+06,
"cpu_time": 2.0910105863874084e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5461112293164559e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8474187014038744e+06,
"cpu_time": 2.0911541701570493e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5474395774268433e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time_median",
"family_index": 6,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8481089780320465e+06,
"cpu_time": 2.0910105863874082e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5461112293164559e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0110023162121311e+03,
"cpu_time": 3.6397903410918798e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.8646401049289183e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.0885471250691287e-03,
"cpu_time": 1.7405652787515543e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.0894167527251185e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"family_index": 6,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 383,
"real_time": 1.8476445676425141e+06,
"cpu_time": 2.1036261749345884e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0940051076620311e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"family_index": 6,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 383,
"real_time": 1.8416928011448642e+06,
"cpu_time": 2.0975615796344471e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.1169306802155510e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"family_index": 6,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 383,
"real_time": 1.8494914593746490e+06,
"cpu_time": 2.1029704804178430e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0869210742026418e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"family_index": 6,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 383,
"real_time": 1.8409780144123677e+06,
"cpu_time": 2.0982375770234512e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.1196939330010206e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"family_index": 6,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 383,
"real_time": 1.8483944312917364e+06,
"cpu_time": 2.1027481540467693e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0911271848185197e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8456402547732263e+06,
"cpu_time": 2.1010287932114201e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.1017355959799528e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time_median",
"family_index": 6,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8476445676425141e+06,
"cpu_time": 2.1027481540467693e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.0940051076620311e+07,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.9922866438279293e+03,
"cpu_time": 2.8846663801736222e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.5371738239379437e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.1630903603792827e-03,
"cpu_time": 1.3729780331874527e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.1645044414327179e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"family_index": 6,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 381,
"real_time": 1.8437174088511069e+06,
"cpu_time": 2.1220168188973828e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4218230990363771e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"family_index": 6,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 381,
"real_time": 1.8506196888653792e+06,
"cpu_time": 2.1262741758530764e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4165201071686497e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"family_index": 6,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 381,
"real_time": 1.8494744966780704e+06,
"cpu_time": 2.1247885800524573e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4173972145647284e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"family_index": 6,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 381,
"real_time": 1.8511147997534547e+06,
"cpu_time": 2.1261485905512171e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4161412357294872e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"family_index": 6,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 381,
"real_time": 1.8506726036777140e+06,
"cpu_time": 2.1259719553803774e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4164796057339334e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8491197995651450e+06,
"cpu_time": 2.1250400241469024e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4176722524466354e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time_median",
"family_index": 6,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8506196888653792e+06,
"cpu_time": 2.1259719553803774e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4165201071686497e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.0802812268848470e+03,
"cpu_time": 1.7904985228232515e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.3663806940135392e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.6658094449095362e-03,
"cpu_time": 8.4257167040514814e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.6692015308401584e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"family_index": 6,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 377,
"real_time": 1.8709263801685872e+06,
"cpu_time": 2.1862395331563847e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.8022909161864346e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"family_index": 6,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 377,
"real_time": 1.8763027216594380e+06,
"cpu_time": 2.1917145809018970e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7942612561810374e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"family_index": 6,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 377,
"real_time": 1.8762896063308965e+06,
"cpu_time": 2.1884845145890196e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7942807881628174e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"family_index": 6,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 377,
"real_time": 1.8757027482789315e+06,
"cpu_time": 2.1878163633952779e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7951550451214367e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"family_index": 6,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 377,
"real_time": 1.8757544488019855e+06,
"cpu_time": 2.1884645570290061e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7950780035993218e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8749951810479679e+06,
"cpu_time": 2.1885439098143172e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7962132018502098e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time_median",
"family_index": 6,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8757544488019857e+06,
"cpu_time": 2.1884645570290061e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7950780035993218e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.2922411942881054e+03,
"cpu_time": 1.9938941407494913e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.4238610091532627e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.2225317789920564e-03,
"cpu_time": 9.1105969215790573e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2244635018845303e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"family_index": 6,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 379,
"real_time": 1.8740940875451188e+06,
"cpu_time": 2.2680971398416646e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5951086285829592e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"family_index": 6,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 379,
"real_time": 1.8682057206275675e+06,
"cpu_time": 2.2638783773087524e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.6127437595457232e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"family_index": 6,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 379,
"real_time": 1.8778234649735866e+06,
"cpu_time": 2.2703807651714310e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5839966831746316e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"family_index": 6,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 379,
"real_time": 1.8769200396875171e+06,
"cpu_time": 2.2739890923481886e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5866844502047849e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"family_index": 6,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 379,
"real_time": 1.8742114155292807e+06,
"cpu_time": 2.2688146596304448e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5947583677686667e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8742509456726140e+06,
"cpu_time": 2.2690320068600965e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5946583778553522e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time_median",
"family_index": 6,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8742114155292804e+06,
"cpu_time": 2.2688146596304448e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.5947583677686667e+08,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.7569226092862182e+03,
"cpu_time": 3.6699100542305432e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.1231084211953892e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.0044928444402990e-03,
"cpu_time": 1.6173901659981393e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.0074655954702278e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"family_index": 6,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 366,
"real_time": 1.9187165077208271e+06,
"cpu_time": 2.4691100874318806e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0929973195941956e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"family_index": 6,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 366,
"real_time": 1.9166553033581972e+06,
"cpu_time": 2.4675474262295594e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0941727478725843e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"family_index": 6,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 366,
"real_time": 1.9170295121402689e+06,
"cpu_time": 2.4675742814207179e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0939591627145236e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"family_index": 6,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 366,
"real_time": 1.9181075183384607e+06,
"cpu_time": 2.4680288797812765e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0933443406846318e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"family_index": 6,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 366,
"real_time": 1.9187868733081173e+06,
"cpu_time": 2.4694711748633790e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0929572372904394e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9178591429731746e+06,
"cpu_time": 2.4683463699453631e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0934861616312749e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time_median",
"family_index": 6,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9181075183384609e+06,
"cpu_time": 2.4680288797812765e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0933443406846318e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 9.7403353919259848e+02,
"cpu_time": 8.9214380256776440e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.5542131957640953e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 5.0787543118656581e-04,
"cpu_time": 3.6143379771596320e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.0793630414839978e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"family_index": 6,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 348,
"real_time": 2.0160473847692171e+06,
"cpu_time": 2.8773849885059744e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0804590366709733e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"family_index": 6,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 348,
"real_time": 2.0210032820321575e+06,
"cpu_time": 2.8825611379309664e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0753573422120061e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"family_index": 6,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 348,
"real_time": 2.0142905912549372e+06,
"cpu_time": 2.8765861321839262e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0822735399795899e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"family_index": 6,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 348,
"real_time": 2.0140643588176274e+06,
"cpu_time": 2.8780579425287396e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0825074341031978e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"family_index": 6,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 348,
"real_time": 2.0120837080062251e+06,
"cpu_time": 2.8772296091952692e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0845574084768763e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0154978649760329e+06,
"cpu_time": 2.8783639620689759e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0810309522885289e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time_median",
"family_index": 6,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0142905912549377e+06,
"cpu_time": 2.8773849885059744e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0822735399795899e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.3830646802996648e+03,
"cpu_time": 2.4039424880061338e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.4884400072009265e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.6785255589143947e-03,
"cpu_time": 8.3517669053852868e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.6763037586560916e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"family_index": 6,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 337,
"real_time": 2.0696820773444076e+06,
"cpu_time": 3.6234149169139820e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0530901300374384e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"family_index": 6,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 337,
"real_time": 2.0703424997379016e+06,
"cpu_time": 3.6245914243321349e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0517972273002996e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"family_index": 6,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 337,
"real_time": 2.0670214165637726e+06,
"cpu_time": 3.6241811869435189e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0583072496390800e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"family_index": 6,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 337,
"real_time": 2.0727762259482141e+06,
"cpu_time": 3.6242709999997290e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0470398564912806e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"family_index": 6,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 337,
"real_time": 2.0653036829272080e+06,
"cpu_time": 3.6237895964393276e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0616825841856875e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0690251805043011e+06,
"cpu_time": 3.6240496249257391e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0543834095307574e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time_median",
"family_index": 6,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0696820773444078e+06,
"cpu_time": 3.6241811869435189e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0530901300374384e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.9198373022969940e+03,
"cpu_time": 4.5561254521534954e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.7219566900003711e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.4112140005881019e-03,
"cpu_time": 1.2571917947306962e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4113013279774182e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"family_index": 6,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 192,
"real_time": 3.6500133173831273e+06,
"cpu_time": 6.6844505572915366e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5964807635355167e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"family_index": 6,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 192,
"real_time": 3.6519199711619876e+06,
"cpu_time": 6.6911448385417545e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5940809580944166e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"family_index": 6,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 192,
"real_time": 3.6467450033039008e+06,
"cpu_time": 6.6815051822916390e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.6006002571608582e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"family_index": 6,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 192,
"real_time": 3.6581350298850643e+06,
"cpu_time": 6.6960100052083274e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5862757560721121e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"family_index": 6,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 192,
"real_time": 3.6469682975924420e+06,
"cpu_time": 6.6835944166664751e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.6003185744925547e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6507563238653047e+06,
"cpu_time": 6.6873409999999478e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5955512618710918e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time_median",
"family_index": 6,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6500133173831268e+06,
"cpu_time": 6.6844505572915366e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.5964807635355167e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.6580916660305502e+03,
"cpu_time": 6.0429536022811353e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.8583358864715155e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.2759251105257858e-03,
"cpu_time": 9.0364071493904421e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2747841450660431e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"family_index": 6,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 97,
"real_time": 7.4139026800151337e+06,
"cpu_time": 1.3824893556700997e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5258797489275255e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"family_index": 6,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 97,
"real_time": 7.4109567050850885e+06,
"cpu_time": 1.3827419412370883e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5276788591918707e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"family_index": 6,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 97,
"real_time": 7.4135843357167291e+06,
"cpu_time": 1.3798558670102756e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5260740932484493e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"family_index": 6,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 97,
"real_time": 7.4144865856649950e+06,
"cpu_time": 1.3862988835052015e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5255233268441544e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"family_index": 6,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 97,
"real_time": 7.4151035340637276e+06,
"cpu_time": 1.3817825010308757e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5251467961110229e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.4136067681091353e+06,
"cpu_time": 1.3826337096907083e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5260605648646040e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time_median",
"family_index": 6,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.4139026800151337e+06,
"cpu_time": 1.3824893556700995e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5258797489275255e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.5910129055633038e+03,
"cpu_time": 2.3402466418205997e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 9.7151325425853045e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.1460713460111087e-04,
"cpu_time": 1.6926005965412973e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.1464875255985290e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"family_index": 6,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 49,
"real_time": 1.4598022633213170e+07,
"cpu_time": 2.6711362836733442e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5971201501849346e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"family_index": 6,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 49,
"real_time": 1.4579789047794683e+07,
"cpu_time": 2.6738295612246048e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6028693405650330e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"family_index": 6,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 49,
"real_time": 1.4619260209099371e+07,
"cpu_time": 2.6772824081634250e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5904418582159081e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"family_index": 6,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 49,
"real_time": 1.4571795582163090e+07,
"cpu_time": 2.6728643653061170e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6053942783925686e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"family_index": 6,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 49,
"real_time": 1.4586992181685505e+07,
"cpu_time": 2.6735653918366350e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6005964193397999e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4591171930791164e+07,
"cpu_time": 2.6737356020408254e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.5992844093396482e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time_median",
"family_index": 6,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4586992181685507e+07,
"cpu_time": 2.6735653918366350e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6005964193397999e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8428013431421845e+04,
"cpu_time": 2.2434985474255573e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.8047034823287912e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.2629563628493712e-03,
"cpu_time": 8.3908765912124063e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2620883958690029e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"family_index": 6,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 25,
"real_time": 2.8088959902524948e+07,
"cpu_time": 5.0339473200001515e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7783089322554455e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"family_index": 6,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 25,
"real_time": 2.8152569979429245e+07,
"cpu_time": 5.0403958599999897e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7675124543894691e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"family_index": 6,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 25,
"real_time": 2.8096582219004631e+07,
"cpu_time": 5.0317505599996366e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7770126257283573e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"family_index": 6,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 25,
"real_time": 2.8130265772342682e+07,
"cpu_time": 5.0390466840003684e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7712925674510031e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"family_index": 6,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 25,
"real_time": 2.8115712553262711e+07,
"cpu_time": 5.0310343319997624e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7737622777916965e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.8116818085312851e+07,
"cpu_time": 5.0352349511999816e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7735777715231953e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time_median",
"family_index": 6,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.8115712553262711e+07,
"cpu_time": 5.0339473200001523e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7737622777916965e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.5718811465242183e+04,
"cpu_time": 4.2605300256099385e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.3653859331060294e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 9.1471273126302684e-04,
"cpu_time": 8.4614324195429681e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 9.1448932897831127e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"family_index": 6,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 10,
"real_time": 5.4613450169563293e+07,
"cpu_time": 9.4311862099993959e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9151894847617989e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"family_index": 6,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 10,
"real_time": 5.4540522024035454e+07,
"cpu_time": 9.4188465599995643e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9217617660810661e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"family_index": 6,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 10,
"real_time": 5.4544457793235779e+07,
"cpu_time": 9.4104660300001800e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9214066260878572e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"family_index": 6,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 10,
"real_time": 4.9170000478625298e+07,
"cpu_time": 9.1761324000003695e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.4593340123454266e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"family_index": 6,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 10,
"real_time": 5.4541722312569618e+07,
"cpu_time": 9.4207825800003782e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9216534538759279e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.3482030555605888e+07,
"cpu_time": 9.3714827559999779e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.0278690686304150e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time_median",
"family_index": 6,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.4541722312569618e+07,
"cpu_time": 9.4188465599995643e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9216534538759279e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.4106957590811700e+06,
"cpu_time": 1.0945291480371244e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.4121227913393396e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 4.5074873448844498e-02,
"cpu_time": 1.1679359355768599e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.7975051824418309e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"family_index": 6,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 8,
"real_time": 1.0172288864850998e+08,
"cpu_time": 1.7617778375000626e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.2777788670068798e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"family_index": 6,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 8,
"real_time": 1.0673330724239349e+08,
"cpu_time": 1.8159297537499696e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0300222664398031e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"family_index": 6,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 8,
"real_time": 1.0171758383512497e+08,
"cpu_time": 1.7591166687500960e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.2780541156995955e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"family_index": 6,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 8,
"real_time": 1.0687513183802366e+08,
"cpu_time": 1.8188923112499821e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0233473659116831e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"family_index": 6,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 8,
"real_time": 1.0180090554058552e+08,
"cpu_time": 1.7678046224999377e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.2737341495057993e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.0376996342092752e+08,
"cpu_time": 1.7847042387500098e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.1765873529127522e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time_median",
"family_index": 6,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.0180090554058552e+08,
"cpu_time": 1.7678046224999377e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.2737341495057993e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.7705344036383042e+06,
"cpu_time": 3.0040796138109593e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.3687270066773024e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.6698808714040307e-02,
"cpu_time": 1.6832366666619164e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.6440720756062384e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"family_index": 6,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 4,
"real_time": 1.8560825288295746e+08,
"cpu_time": 3.1317242525000209e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.7849896614084826e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"family_index": 6,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 4,
"real_time": 1.8542753159999847e+08,
"cpu_time": 3.1282622100002301e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.7906278249783316e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"family_index": 6,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 4,
"real_time": 1.8527872487902641e+08,
"cpu_time": 3.1191100350000054e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.7952785712502909e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"family_index": 6,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 4,
"real_time": 1.8555964156985283e+08,
"cpu_time": 3.1290797925001359e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.7865051630626059e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"family_index": 6,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 4,
"real_time": 1.8527125194668770e+08,
"cpu_time": 3.1231620375001514e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.7955123243241863e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time_mean",
"family_index": 6,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8542908057570460e+08,
"cpu_time": 3.1262676655001086e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.7905827090047798e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time_median",
"family_index": 6,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8542753159999847e+08,
"cpu_time": 3.1282622100002301e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.7906278249783316e+09,
"dst_gpu": 6.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time_stddev",
"family_index": 6,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.5545738954361441e+05,
"cpu_time": 5.0630328698259249e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.8544522649790263e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time_cv",
"family_index": 6,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/6/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 8.3836574641347203e-04,
"cpu_time": 1.6195135578757271e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.3833570970845439e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"family_index": 7,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 384,
"real_time": 1.8392240839564996e+06,
"cpu_time": 2.0731707968749311e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2270260789477988e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"family_index": 7,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 384,
"real_time": 1.8353011443916026e+06,
"cpu_time": 2.0687297552080725e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2317863270105533e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"family_index": 7,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 384,
"real_time": 1.8312123837252632e+06,
"cpu_time": 2.0692872630207285e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2367694956646399e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"family_index": 7,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 384,
"real_time": 1.8339949891128528e+06,
"cpu_time": 2.0673252968750333e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2333757858200767e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"family_index": 7,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 384,
"real_time": 1.8417149730301693e+06,
"cpu_time": 2.0753482421875801e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2240140629691798e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8362895148432776e+06,
"cpu_time": 2.0707722708332692e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2305943500824496e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time_median",
"family_index": 7,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8353011443916026e+06,
"cpu_time": 2.0692872630207285e+06,
"time_unit": "ns",
"bytes": 4.0960000000000000e+03,
"bytes_per_second": 2.2317863270105533e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.1852800767488798e+03,
"cpu_time": 3.3522823851744602e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 5.0823802028785021e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 0,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):12/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.2792049090941320e-03,
"cpu_time": 1.6188561303390047e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.2784869883179979e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"family_index": 7,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 384,
"real_time": 1.8353736700191803e+06,
"cpu_time": 2.0694872708334106e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4633962739121076e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"family_index": 7,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 384,
"real_time": 1.8286399621464016e+06,
"cpu_time": 2.0680846145832988e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4798320990341259e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"family_index": 7,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 384,
"real_time": 1.8302132919908825e+06,
"cpu_time": 2.0696319713543095e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4759810432197489e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"family_index": 7,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 384,
"real_time": 1.8339162383161541e+06,
"cpu_time": 2.0676207760415012e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4669433798795762e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"family_index": 7,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 384,
"real_time": 1.8386333635286670e+06,
"cpu_time": 2.0728169088541174e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4554831661914820e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8333553052002571e+06,
"cpu_time": 2.0695283083333275e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4683271924474081e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time_median",
"family_index": 7,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8339162383161546e+06,
"cpu_time": 2.0694872708334110e+06,
"time_unit": "ns",
"bytes": 8.1920000000000000e+03,
"bytes_per_second": 4.4669433798795762e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.0110176184860552e+03,
"cpu_time": 2.0341002048977200e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 9.7742290294486756e+03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 1,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):13/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.1878015718551250e-03,
"cpu_time": 9.8288107328952691e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.1874470262539347e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"family_index": 7,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 384,
"real_time": 1.8346516044213483e+06,
"cpu_time": 2.0704578723957923e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.9303058741594367e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"family_index": 7,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 384,
"real_time": 1.8307999760433326e+06,
"cpu_time": 2.0714236640625808e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.9490934096517656e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"family_index": 7,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 384,
"real_time": 1.8309274576608913e+06,
"cpu_time": 2.0712248281249402e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.9484703129262384e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"family_index": 7,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 384,
"real_time": 1.8353020432944808e+06,
"cpu_time": 2.0707376406251162e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.9271409356629420e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"family_index": 7,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 384,
"real_time": 1.8401604312051253e+06,
"cpu_time": 2.0756572369794075e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.9035715159194469e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8343683025250356e+06,
"cpu_time": 2.0719002484375674e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.9317164096639659e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time_median",
"family_index": 7,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8346516044213485e+06,
"cpu_time": 2.0712248281249399e+06,
"time_unit": "ns",
"bytes": 1.6384000000000000e+04,
"bytes_per_second": 8.9303058741594367e+06,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.8429563390148032e+03,
"cpu_time": 2.1348532584980621e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.8693000158920619e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 2,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):14/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.0949753295043945e-03,
"cpu_time": 1.0303841896384578e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.0928788265931852e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"family_index": 7,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 384,
"real_time": 1.8379525645097299e+06,
"cpu_time": 2.0765343749999944e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7828534116026439e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"family_index": 7,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 384,
"real_time": 1.8319450124219356e+06,
"cpu_time": 2.0765649296876632e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7886999761351377e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"family_index": 7,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 384,
"real_time": 1.8336800533613011e+06,
"cpu_time": 2.0770974895834017e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7870074956606142e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"family_index": 7,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 384,
"real_time": 1.8375908928192074e+06,
"cpu_time": 2.0763819817706991e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7832043099499568e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"family_index": 7,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 384,
"real_time": 1.8425534165089631e+06,
"cpu_time": 2.0813165781250475e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7784016303898890e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8367443879242279e+06,
"cpu_time": 2.0775790708333612e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7840333647476483e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time_median",
"family_index": 7,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8375908928192076e+06,
"cpu_time": 2.0765649296876632e+06,
"time_unit": "ns",
"bytes": 3.2768000000000000e+04,
"bytes_per_second": 1.7832043099499568e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.1335585703741262e+03,
"cpu_time": 2.1067614439621802e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.0130652152531737e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 3,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):15/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.2504811216794364e-03,
"cpu_time": 1.0140463357272429e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.2494339481262011e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"family_index": 7,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 383,
"real_time": 1.8386862199089683e+06,
"cpu_time": 2.0829171801564700e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5642840681780182e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"family_index": 7,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 383,
"real_time": 1.8343728774074614e+06,
"cpu_time": 2.0835201409919818e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5726651220782720e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"family_index": 7,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 383,
"real_time": 1.8353993087733716e+06,
"cpu_time": 2.0850296631853385e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5706671396644913e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"family_index": 7,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 383,
"real_time": 1.8392639762900376e+06,
"cpu_time": 2.0832539765013510e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5631644421260320e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"family_index": 7,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 383,
"real_time": 1.8457254157346911e+06,
"cpu_time": 2.0896987075719382e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5506906629398823e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8386895596229057e+06,
"cpu_time": 2.0848839336814159e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5642942869973391e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time_median",
"family_index": 7,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8386862199089683e+06,
"cpu_time": 2.0835201409919816e+06,
"time_unit": "ns",
"bytes": 6.5536000000000000e+04,
"bytes_per_second": 3.5642840681780182e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.4523233509777619e+03,
"cpu_time": 2.8101643653205456e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.6171249909337508e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 4,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):16/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.4214655093222383e-03,
"cpu_time": 1.3478756874290144e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.4176244431806044e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"family_index": 7,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 383,
"real_time": 1.8460500051705751e+06,
"cpu_time": 2.1031980339425984e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.1001326959119365e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"family_index": 7,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 383,
"real_time": 1.8409099172622659e+06,
"cpu_time": 2.0984541462141052e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.1199572977979004e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"family_index": 7,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 383,
"real_time": 1.8457955987859573e+06,
"cpu_time": 2.1029335953000896e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.1011113086525142e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"family_index": 7,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 383,
"real_time": 1.8367323625572002e+06,
"cpu_time": 2.0976406892951760e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.1361512799564511e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"family_index": 7,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 383,
"real_time": 1.8449120524287096e+06,
"cpu_time": 2.1023949582244596e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.1045121000457466e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8428799872409417e+06,
"cpu_time": 2.1009242845952855e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.1123729364729106e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time_median",
"family_index": 7,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8449120524287098e+06,
"cpu_time": 2.1023949582244596e+06,
"time_unit": "ns",
"bytes": 1.3107200000000000e+05,
"bytes_per_second": 7.1045121000457466e+07,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.0110334782123814e+03,
"cpu_time": 2.6577118089620135e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.5503004990565538e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 5,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):17/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 2.1765028140641319e-03,
"cpu_time": 1.2650202715296737e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.1797232975600427e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"family_index": 7,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 381,
"real_time": 1.8497281179982151e+06,
"cpu_time": 2.1260759317586059e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4172028713262656e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"family_index": 7,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 381,
"real_time": 1.8496760646516231e+06,
"cpu_time": 2.1262214304460655e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4172427540677154e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"family_index": 7,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 381,
"real_time": 1.8495215221660777e+06,
"cpu_time": 2.1259530000000405e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4173611761651117e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"family_index": 7,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 381,
"real_time": 1.8495080862796237e+06,
"cpu_time": 2.1261073569555297e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4173714726888031e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"family_index": 7,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 381,
"real_time": 1.8482024730149573e+06,
"cpu_time": 2.1244876482940651e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4183727369023952e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8493272528220997e+06,
"cpu_time": 2.1257690734908609e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4175102022300583e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time_median",
"family_index": 7,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8495215221660777e+06,
"cpu_time": 2.1260759317586059e+06,
"time_unit": "ns",
"bytes": 2.6214400000000000e+05,
"bytes_per_second": 1.4173611761651117e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 6.3599007668374281e+02,
"cpu_time": 7.2268665647704006e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.8769906653591206e+04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 6,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):18/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 3.4390347933996697e-04,
"cpu_time": 3.3996479932331977e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.4405330259256909e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"family_index": 7,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 376,
"real_time": 1.8745907105535190e+06,
"cpu_time": 2.1887260797869693e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7968131765957117e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"family_index": 7,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 376,
"real_time": 1.8736316541957848e+06,
"cpu_time": 2.1878461861703917e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.7982447821369624e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"family_index": 7,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 376,
"real_time": 1.8712703040751449e+06,
"cpu_time": 2.1884626835105671e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.8017758784406275e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"family_index": 7,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 376,
"real_time": 1.8690281815113539e+06,
"cpu_time": 2.1853959494680637e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.8051369432859194e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"family_index": 7,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 376,
"real_time": 1.8693422704366620e+06,
"cpu_time": 2.1850824069150453e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.8046656211199403e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8715726241544927e+06,
"cpu_time": 2.1871026611702074e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.8013272803158325e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time_median",
"family_index": 7,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8712703040751449e+06,
"cpu_time": 2.1878461861703913e+06,
"time_unit": "ns",
"bytes": 5.2428800000000000e+05,
"bytes_per_second": 2.8017758784406275e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.4944194578583838e+03,
"cpu_time": 1.7343781513876831e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.7329016515306162e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 7,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):19/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.3327933020954878e-03,
"cpu_time": 7.9300262497038231e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.3325474955249621e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"family_index": 7,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 378,
"real_time": 1.8632101904749922e+06,
"cpu_time": 2.2627333994709835e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.6277923197311652e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"family_index": 7,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 378,
"real_time": 1.8698438682818846e+06,
"cpu_time": 2.2691242804231602e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.6078265024527919e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"family_index": 7,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 378,
"real_time": 1.8631861151396863e+06,
"cpu_time": 2.2631281613758132e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.6278650397809899e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"family_index": 7,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 378,
"real_time": 1.8710337567547206e+06,
"cpu_time": 2.2690301375661041e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.6042601915357149e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"family_index": 7,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 378,
"real_time": 1.8654701076066070e+06,
"cpu_time": 2.2637098862432856e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.6209745507276988e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8665488076515782e+06,
"cpu_time": 2.2655451730158692e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.6177437208456731e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time_median",
"family_index": 7,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.8654701076066073e+06,
"cpu_time": 2.2637098862432856e+06,
"time_unit": "ns",
"bytes": 1.0485760000000000e+06,
"bytes_per_second": 5.6209745507276988e+08,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6942472974828715e+03,
"cpu_time": 3.2431189085995284e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.1113173281470959e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 8,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):20/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.9791860155700053e-03,
"cpu_time": 1.4314960245450869e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.9782271733460323e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"family_index": 7,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 367,
"real_time": 1.9078657352390930e+06,
"cpu_time": 2.4622459073568876e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0992136193153999e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"family_index": 7,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 367,
"real_time": 1.9099740909440417e+06,
"cpu_time": 2.4634630953678596e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0980002346332572e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"family_index": 7,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 367,
"real_time": 1.9108743326895812e+06,
"cpu_time": 2.4629891825613142e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0974829501468213e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"family_index": 7,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 367,
"real_time": 1.9102726864110809e+06,
"cpu_time": 2.4636167929158215e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0978286057892697e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"family_index": 7,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 367,
"real_time": 1.9087389648234730e+06,
"cpu_time": 2.4617291580382055e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0987107397338388e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9095451620214540e+06,
"cpu_time": 2.4628088272480173e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0982472299237173e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time_median",
"family_index": 7,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9099740909440417e+06,
"cpu_time": 2.4629891825613142e+06,
"time_unit": "ns",
"bytes": 2.0971520000000000e+06,
"bytes_per_second": 1.0980002346332572e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.2196715402702569e+03,
"cpu_time": 8.0560035590864686e+02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.0162778918740095e+05,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 9,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):21/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 6.3872358953747201e-04,
"cpu_time": 3.2710632956795016e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.3886142397658041e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"family_index": 7,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 351,
"real_time": 2.0102136735233911e+06,
"cpu_time": 2.8781618091170588e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0864966024474685e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"family_index": 7,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 351,
"real_time": 2.0082120278165636e+06,
"cpu_time": 2.8730055270656822e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0885762767591197e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"family_index": 7,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 351,
"real_time": 2.0067569964891695e+06,
"cpu_time": 2.8723533903134754e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0900906324671862e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"family_index": 7,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 351,
"real_time": 2.0073445887302191e+06,
"cpu_time": 2.8741099059827859e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0894788187080429e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"family_index": 7,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 351,
"real_time": 2.0070774304916675e+06,
"cpu_time": 2.8731373675216213e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0897569452378001e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0079209434102022e+06,
"cpu_time": 2.8741536000001249e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0888798551239235e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time_median",
"family_index": 7,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0073445887302191e+06,
"cpu_time": 2.8731373675216213e+06,
"time_unit": "ns",
"bytes": 4.1943040000000000e+06,
"bytes_per_second": 2.0894788187080429e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.3910259432335852e+03,
"cpu_time": 2.3269725137690302e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.4462087818292351e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 10,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):22/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 6.9276927849116501e-04,
"cpu_time": 8.0962009607591235e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.9233698543348643e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"family_index": 7,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 339,
"real_time": 2.0649731907839708e+06,
"cpu_time": 3.6208482064896743e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0623326430767121e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"family_index": 7,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 339,
"real_time": 2.0649514870531305e+06,
"cpu_time": 3.6227892802360379e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0623753403385224e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"family_index": 7,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 339,
"real_time": 2.0608287617141125e+06,
"cpu_time": 3.6214728584070429e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0705021959333982e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"family_index": 7,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 339,
"real_time": 2.0693932745089221e+06,
"cpu_time": 3.6234743746312754e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0536557759861574e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"family_index": 7,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 339,
"real_time": 2.0633495871560059e+06,
"cpu_time": 3.6224496519174604e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0655292017492490e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0646992602432284e+06,
"cpu_time": 3.6222068743362986e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0628790314168077e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time_median",
"family_index": 7,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.0649514870531305e+06,
"cpu_time": 3.6224496519174604e+06,
"time_unit": "ns",
"bytes": 8.3886080000000000e+06,
"bytes_per_second": 4.0623753403385224e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.1214901303798465e+03,
"cpu_time": 1.0475509876262365e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.1391515236032410e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 11,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):23/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.5118376755809582e-03,
"cpu_time": 2.8920241829594027e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.5110347800491601e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"family_index": 7,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 192,
"real_time": 3.6381991036857166e+06,
"cpu_time": 6.6759996458335044e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.6114067762271891e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"family_index": 7,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 192,
"real_time": 3.6357265998958610e+06,
"cpu_time": 6.6752971302085714e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.6145427988123627e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"family_index": 7,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 192,
"real_time": 3.6399115548798968e+06,
"cpu_time": 6.6772452135417387e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.6092372704791126e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"family_index": 7,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 192,
"real_time": 3.6382440885063261e+06,
"cpu_time": 6.6725962291661985e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.6113497588029757e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"family_index": 7,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 192,
"real_time": 3.6385982760596867e+06,
"cpu_time": 6.6792597708330937e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.6109008819100513e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6381359246054976e+06,
"cpu_time": 6.6760795979166217e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.6114874972463379e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time_median",
"family_index": 7,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.6382440885063261e+06,
"cpu_time": 6.6759996458335053e+06,
"time_unit": "ns",
"bytes": 1.6777216000000000e+07,
"bytes_per_second": 4.6113497588029757e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.5155170987029474e+03,
"cpu_time": 2.4609279476111342e+03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.9214573562377074e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 12,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):24/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 4.1656417740007420e-04,
"cpu_time": 3.6861872473466402e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.1666758445839202e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"family_index": 7,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 96,
"real_time": 7.3817636924407752e+06,
"cpu_time": 1.3777454218749577e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5455846865378656e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"family_index": 7,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 96,
"real_time": 7.3868886926599471e+06,
"cpu_time": 1.3756473593749283e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5424309741314611e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"family_index": 7,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 96,
"real_time": 7.3819803074002266e+06,
"cpu_time": 1.3755504177083822e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5454513020527334e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"family_index": 7,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 96,
"real_time": 7.3958270368166268e+06,
"cpu_time": 1.3817279052084075e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5369411470772820e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"family_index": 7,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 96,
"real_time": 7.3892036501395823e+06,
"cpu_time": 1.3808806906250007e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5410078797011032e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.3871326758914320e+06,
"cpu_time": 1.3783103589583356e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5422831979000893e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time_median",
"family_index": 7,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 7.3868886926599471e+06,
"cpu_time": 1.3777454218749579e+07,
"time_unit": "ns",
"bytes": 3.3554432000000000e+07,
"bytes_per_second": 4.5424309741314611e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.8162113617005243e+03,
"cpu_time": 2.8859237532946678e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.5749997369734170e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 13,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):25/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 7.8734356304202427e-04,
"cpu_time": 2.0938127139055371e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.8704906348114748e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"family_index": 7,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 52,
"real_time": 1.4487662943653189e+07,
"cpu_time": 2.6575477884613555e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6321386866195221e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"family_index": 7,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 52,
"real_time": 1.4534653952488532e+07,
"cpu_time": 2.6659296750000734e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6171628316276522e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"family_index": 7,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 52,
"real_time": 1.4473770673458394e+07,
"cpu_time": 2.6542975788458988e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6365847237763968e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"family_index": 7,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 52,
"real_time": 1.4485672223739900e+07,
"cpu_time": 2.6572343769229379e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6327752667230988e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"family_index": 7,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 52,
"real_time": 1.4505435354434527e+07,
"cpu_time": 2.6578438269229293e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6264632780900173e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4497439029554909e+07,
"cpu_time": 2.6585706492306393e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6290249573673372e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time_median",
"family_index": 7,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.4487662943653191e+07,
"cpu_time": 2.6575477884613559e+07,
"time_unit": "ns",
"bytes": 6.7108864000000000e+07,
"bytes_per_second": 4.6321386866195221e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.3680753552916212e+04,
"cpu_time": 4.3524283246113118e+04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.5528541967762094e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 14,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):26/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.6334439141037205e-03,
"cpu_time": 1.6371309620343757e-03,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.6316296123561494e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"family_index": 7,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 25,
"real_time": 2.8028325513005257e+07,
"cpu_time": 5.0096969760006689e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7886459695111799e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"family_index": 7,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 25,
"real_time": 2.8012460097670555e+07,
"cpu_time": 5.0064701999999665e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7913581146398926e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"family_index": 7,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 25,
"real_time": 2.8027845397591591e+07,
"cpu_time": 5.0079308160002261e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7887279987470322e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"family_index": 7,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 25,
"real_time": 2.7254094928503036e+07,
"cpu_time": 4.8781048399996512e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.9246811663384809e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"family_index": 7,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 25,
"real_time": 2.8017516061663628e+07,
"cpu_time": 5.0031099320003703e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7904934793148966e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.7868048399686813e+07,
"cpu_time": 4.9810625528001770e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.8167813457102966e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time_median",
"family_index": 7,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 2.8017516061663628e+07,
"cpu_time": 5.0064701999999672e+07,
"time_unit": "ns",
"bytes": 1.3421772800000000e+08,
"bytes_per_second": 4.7904934793148966e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 3.4327759133939643e+05,
"cpu_time": 5.7605902150586655e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 6.0329001232821353e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 15,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):27/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 1.2317963081449732e-02,
"cpu_time": 1.1564982679890791e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 1.2524753959726413e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"family_index": 7,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 10,
"real_time": 5.4460106790065765e+07,
"cpu_time": 9.3962407300000459e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9290291889212036e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"family_index": 7,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 10,
"real_time": 5.4425880685448647e+07,
"cpu_time": 9.3926933800003096e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9321288442057152e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"family_index": 7,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 10,
"real_time": 5.3398819267749786e+07,
"cpu_time": 9.1091311800005317e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 5.0269923507863321e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"family_index": 7,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 10,
"real_time": 5.4418105259537697e+07,
"cpu_time": 9.3857332199991107e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9328335619137001e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"family_index": 7,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 10,
"real_time": 5.4504762962460518e+07,
"cpu_time": 9.3988104699997157e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9249907973158531e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.4241534993052483e+07,
"cpu_time": 9.3365217959999427e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9491949486285610e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time_median",
"family_index": 7,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 5.4425880685448647e+07,
"cpu_time": 9.3926933800003096e+07,
"time_unit": "ns",
"bytes": 2.6843545600000000e+08,
"bytes_per_second": 4.9321288442057152e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.7233271676589752e+05,
"cpu_time": 1.2721019351095692e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 4.3600118931155220e+07,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 16,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):28/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 8.7079526201903432e-03,
"cpu_time": 1.3625009001259732e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 8.8095375881762274e-03,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"family_index": 7,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 8,
"real_time": 1.0650852415710688e+08,
"cpu_time": 1.8097749162501487e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0406379794360981e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"family_index": 7,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 8,
"real_time": 9.9188854917883873e+07,
"cpu_time": 1.7336862999999881e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.4126132663237505e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"family_index": 7,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 8,
"real_time": 1.0663040727376938e+08,
"cpu_time": 1.8130245399999014e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0348763146107569e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"family_index": 7,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 8,
"real_time": 9.9353916011750698e+07,
"cpu_time": 1.7308065449998367e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.4036210503922529e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"family_index": 7,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 8,
"real_time": 1.0659786686301231e+08,
"cpu_time": 1.8124795037499553e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0364132772931242e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.0365591384470463e+08,
"cpu_time": 1.7799543609999660e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.1856323776111975e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time_median",
"family_index": 7,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.0650852415710688e+08,
"cpu_time": 1.8097749162501484e+08,
"time_unit": "ns",
"bytes": 5.3687091200000000e+08,
"bytes_per_second": 5.0406379794360981e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 4.0031826440448249e+06,
"cpu_time": 4.3580467103129514e+06,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 2.0313572509215173e+08,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 17,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):29/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 3.8619915599242306e-02,
"cpu_time": 2.4484036252843198e-02,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.9172797124837413e-02,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"family_index": 7,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 0,
"threads": 1,
"iterations": 4,
"real_time": 1.9507104530930519e+08,
"cpu_time": 3.1270679149997705e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.5043628965922232e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"family_index": 7,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 1,
"threads": 1,
"iterations": 4,
"real_time": 1.9528680667281151e+08,
"cpu_time": 3.1206743174999475e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.4982814368969345e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"family_index": 7,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 2,
"threads": 1,
"iterations": 4,
"real_time": 1.9541329517960548e+08,
"cpu_time": 3.1276148174998754e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.4947224702040758e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"family_index": 7,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 3,
"threads": 1,
"iterations": 4,
"real_time": 1.9522952660918236e+08,
"cpu_time": 3.1263332399998945e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.4998946247995358e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"family_index": 7,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"run_type": "iteration",
"repetitions": 5,
"repetition_index": 4,
"threads": 1,
"iterations": 4,
"real_time": 1.9538485631346703e+08,
"cpu_time": 3.1267374200001544e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.4955222439416437e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time_mean",
"family_index": 7,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "mean",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9527710601687431e+08,
"cpu_time": 3.1256855419999284e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.4985567344868822e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time_median",
"family_index": 7,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "median",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.9528680667281151e+08,
"cpu_time": 3.1267374200001538e+08,
"time_unit": "ns",
"bytes": 1.0737418240000000e+09,
"bytes_per_second": 5.4982814368969345e+09,
"dst_gpu": 7.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time_stddev",
"family_index": 7,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "stddev",
"aggregate_unit": "time",
"iterations": 5,
"real_time": 1.3690921482242667e+05,
"cpu_time": 2.8403956623203168e+05,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 3.8563899288324048e+06,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": 0.0000000000000000e+00
},
{
"name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time_cv",
"family_index": 7,
"per_family_instance_index": 18,
"run_name": "Comm_implicit_managed_GPURdHost_fine/0/7/log2(N):30/manual_time",
"run_type": "aggregate",
"repetitions": 5,
"threads": 1,
"aggregate_name": "cv",
"aggregate_unit": "percentage",
"iterations": 5,
"real_time": 7.0110223166967682e-04,
"cpu_time": 9.0872726131718525e-04,
"time_unit": "ns",
"bytes": 0.0000000000000000e+00,
"bytes_per_second": 7.0134584674650584e-04,
"dst_gpu": 0.0000000000000000e+00,
"src_numa": NaN
}
]
}