Create a NCCL sub-communicator using ncclCommSplit #2495
+36
−53
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
seunghwak
added
improvement
seunghwak
changed the title
|
@seunghwak is this PR ready for review? Is this something you'd like to include in the 25.02 release? I Just changed the base branch accordingly. Hopefully I didn't mess anything up. |
cjnolet
approved these changes
Jan 28, 2025
|
Yes, this PR is ready to review but it seems like too late for 25.02, let's target 25.04 |
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Outdated suggestions cannot be applied.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.
NCCL expects to create a sub-communicator using
ncclCommSplit(like we create a MPI sub-communicator usingMPI_Comm_split) but raft currently creates a sub-communicator from scratch usingncclCommInitRankon a subset of GPUs.This first prevents sharing resources among the global communicator and sub-communicators. We also experienced erroneous sub-communicator behaviors when we have a non-power-of-two number of GPUs.
This PR updates raft to create NCCL sub-communicators using
ncclCommSplit.