[CPU] Faster reduce kernel for SHM allreduce #4049
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
delock
requested review from
jeffra,
RezaYazdaniAminabadi,
cmikeh2,
awan-10 and
arashb
as code owners
tjruwase
approved these changes
Jul 26, 2023
polisettyvarma
pushed a commit
to polisettyvarma/DeepSpeed
that referenced
this pull request
Aug 7, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
This PR make reduce kernel in SHM based allreduce faster by using
#pragma omp parallel for.On a server with 2 socket and SNC4 (total 8 ranks), the allreduce time can reduce from 15us to 12us for 10KB message size, and 100us to 60us for 100KB message size.