llnl / llnl/RAJAPerf

MAT_MAT_SHARED Tuning

Open
#534 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Jupyter Notebook
Stars
135
Forks
55
Avg merge
4d 16h
Merged PRs (30d)
5

Description

Try doing a tuning of MAT_MAT_SHARED where we run multiple blocks per output tile and atomically increment the output tile.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by locating the MAT_MAT_SHARED implementation and its associated performance or correctness checks in RAJAPerf. Read how output tiles are currently computed, then investigate the requested multiple-blocks-per-output-tile and atomic-increment approach. Done means the tuning is implemented for MAT_MAT_SHARED and its existing correctness and performance checks still pass.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
performance
Issue type
Refactor
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.