prevent saving a large number of small files#78
Open
LiuXTao wants to merge 1 commit intoISEEKYAN:mainfrom
Open
prevent saving a large number of small files#78LiuXTao wants to merge 1 commit intoISEEKYAN:mainfrom
LiuXTao wants to merge 1 commit intoISEEKYAN:mainfrom
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Originally, the
_save_weight_fastfunction saved each small weight as an individual file. When the number of weights is large, this results in a concentrated burst of creating and deleting a large number of new files in a short period. This not only may put pressure on the distributed file system but is also relatively inefficient. Therefore, I have added a new logic for saving batch files here.I have verified the correctness, and testing before and after the modifications showed that the
save_weightstime for an 80B MoE model on 16 GPUs was reduced from 250s to 190s, a decrease of 24%.