Improve CCL performance #298
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Motivation
CCL Performance was not equivalent to RCCL
Technical Details
Improve CCL kernels to bring performance closer to parity with rccl and add a comprehensive benchmarking script for ccl.
Test Plan
Compare against RCCL's collective
Test Result
iris.ccl was competitive on the MI350 machine I was running on
Submission Checklist