17 lines
1.3 KiB
Markdown
17 lines
1.3 KiB
Markdown
---
|
|
title: Automatic Discovery of Implementation Rules for Fast GPU + MPI Operations
|
|
date: "2022-02-22T00:00:00Z"
|
|
tags: [MPI, CUDA]
|
|
---
|
|
|
|
*Feb 25th, 4:00 PM, MS61: Experiences in Developing GPU Support for DOE Math Libraries*
|
|
|
|
Invited talk to SIAM Parallel Processing Mini-symposium
|
|
|
|
### Abstract
|
|
|
|
Developing a high-performance implementation of a distributed computational kernel for high-performacing computing is increasingly challenging. Systems are composed of heterogenous computational resources, and limited communication performance demands an asynchronous application design. Even if high-performance computation and communication libraries are available. the challenge becomes the best coordination of the provided operations to create an optimal result. This work presents a system that automatically generates design rules for a high-performance implementation of a compound operation provided as a dependence graph. The system searches among valid schedules to determine the fastest arrangement of operations. A post-processing step on the results of the search yields interpretable design rules. The fast implementation can be used directly, or experts can use the design rules to create a high-performance implementation.
|
|
|
|
### Link
|
|
|
|
* [slides](/pdf/20220225_siampp.pdf) |