mirror of
https://github.com/deepseek-ai/DeepEP.git
synced 2026-11-05 00:47:32 +00:00
[GH-ISSUE #455] Maybe use reference stash to replace record stream to reduce mem peak #296
Labels
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
deepseek-ai/DeepEP#296
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @junjzhang on GitHub (Oct 15, 2025).
Original GitHub issue: https://github.com/deepseek-ai/DeepEP/issues/455
Due to behavior of CudaCacheAllocator, record stream will lead to a late memory free, which has significant on memory peak. (Could refer to fsdp1's issue due to recordstream). In Pytorch 2.8 +，c10d has remove all recordStream in collective communication. Thus, I wonder if there are plans to remove record stream and use reference stash to handle multistream scenario?
@junjzhang commented on GitHub (Oct 15, 2025):
https://github.com/deepseek-ai/DeepEP/pull/456 A tested PR, could be further discussed.