24 lines
1.9 KiB
Markdown
24 lines
1.9 KiB
Markdown
---
|
|
title: Automatic Discovery of Implementation Rules for Fast GPU + MPI Operations
|
|
date: "2022-02-25T00:00:00Z"
|
|
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.
|
|
|
|
### Video
|
|
|
|
<div style="padding:56.25% 0 0 0;position:relative;"><iframe src="https://player.vimeo.com/video/682527688?h=88d35e0542&badge=0&autopause=0&player_id=0&app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen style="position:absolute;top:0;left:0;width:100%;height:100%;" title="Automatic Discovery of Implementation Rules for Fast GPU + MPI Operations"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>
|
|
|
|
### Link
|
|
|
|
* [slides](/pdf/20220225_siampp.pdf)
|
|
* [github (tenzing-core)](https://github.com/sandialabs/tenzing-core)
|
|
* [github (tenzing-mcts)](https://github.com/sandialabs/tenzing-mcts)
|
|
* [github (tenzing-dfs)](https://github.com/sandialabs/tenzing-dfs) |