Multi-node AllReduce tree/ring algo switching fix for MI300 #1624
+4
−4
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Outdated suggestions cannot be applied.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.
Details
Do not mention proprietary info or link to internal work items in this PR.
Work item: Internal
What were the changes?
Adjust tree/ring correction factor to address the AllReduce Tree->Ring switching point for large messages (above 64 MB) and get a performance boost up to 1.25x for the impacted message sizes
Why were the changes made?
Suboptimal tree usage for larger messages where ring works better
How was the outcome achieved?
Tune correction factors for MI300x and use topo_explorer to mimic the real runs
Additional Documentation:
Approval Checklist
Do not approve until these items are satisfied.