Add reduce-scatter coalescing for FSDP/ZeRO1#5938
Closed
jeffhataws wants to merge 7 commits intopytorch:masterfrom
Closed
Add reduce-scatter coalescing for FSDP/ZeRO1#5938jeffhataws wants to merge 7 commits intopytorch:masterfrom
jeffhataws wants to merge 7 commits intopytorch:masterfrom
Conversation
Also allow using reduce-scatter's scale param in FSDP. (revived pytorch#4145)
…ter tuple change without token
Collaborator
|
Build seems to fail with oom, let's use a larger machine. can you modify https://github.com/pytorch/xla/blob/master/.github/workflows/build_and_test.yml#L22 in your pr and add I think we used larger machine to run the test, but use default machine to build. |
This was referenced Nov 30, 2023
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
(Replaced by #5956)
This PR adds reduce-scatter coalescence support and use that in FSDP/ZeRO1. This also enables using reduce-scatter's scale param in FSD.. This PR is companion to #5624 and to be used in conjunction with openxla openxla/xla#5740 .
This is a revival of #4145 . Will need to address the comments.