Refactor foreach allgather #1416
Open
Conversation
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
CyCle1024
force-pushed
the
refactor_foreach_allgather
branch
from
8fbecd5 to
92e40de
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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
In the case of
len(params)==1,foreach_all_gatheris in low performance due to copyin operation.Besides, in the case of
_get_fused_paramsusage, local tensor of params are padded to the same size on each rank, this behaviour can avoid one all gather communication of num elements.Key Changes
params_shapes_across_groupparameter inforeach_all_gather, it aims to provide static shape information for theforeach_all_gatherrather than all gather num elements metadata.len(params)==1offoreach_all_gather, avoiding copyin operation and copyout metadata all gather. It's compatible whether we provide theparams_shapes_across_groupor not.