Skip to content

Conversation

EddyLXJ
Copy link
Contributor

@EddyLXJ EddyLXJ commented Oct 8, 2025

Summary:
Before KVZCH is using ID_COUNT and MEM_UTIL eviction trigger mode, both are very tricky and hard for model engineer to decide what num to use for the id count or mem util threshold. Besides that, the eviction start time is out of sync after some time in training, which can cause great qps drop during eviction.

This diff is adding support for free memory trigger eviction. It will check how many free memory left every N batch in every rank and if free memory below the threshold, it will trigger eviction in all tbes of all ranks using all reduce. In this way, we can force the start time of eviction in all ranks.

Differential Revision: D83896528

@meta-cla meta-cla bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Oct 8, 2025
Copy link
Contributor

meta-codesync bot commented Oct 8, 2025

@EddyLXJ has exported this pull request. If you are a Meta employee, you can view the originating Diff in D83896528.

…#3442)

Summary:

Before KVZCH is using ID_COUNT and MEM_UTIL eviction trigger mode, both are very tricky and hard for model engineer to decide what num to use for the id count or mem util threshold. Besides that, the eviction start time is out of sync after some time in training, which can cause great qps drop during eviction. 

This diff is adding support for free memory trigger eviction. It will check how many free memory left every N batch in every rank and if free memory below the threshold, it will trigger eviction in all tbes of all ranks using all reduce. In this way, we can force the start time of eviction in all ranks.

Differential Revision: D83896528
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. fb-exported meta-exported
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant