Free mem trigger with all2all for sync trigger eviction #3490
+34
−11
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
Contributor
meta-cla
bot
added
the
CLA Signed
EddyLXJ
added a commit
to EddyLXJ/FBGEMM-1
that referenced
this pull request
Oct 27, 2025
Summary: X-link: meta-pytorch/torchrec#3490 X-link: facebookresearch/FBGEMM#2070 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: D85604160
EddyLXJ
added a commit
to EddyLXJ/torchrec
that referenced
this pull request
Oct 27, 2025
…#3490) Summary: X-link: pytorch/FBGEMM#5062 X-link: facebookresearch/FBGEMM#2070 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: D85604160
EddyLXJ
added a commit
to EddyLXJ/FBGEMM-1
that referenced
this pull request
Oct 27, 2025
Summary: X-link: meta-pytorch/torchrec#3490 X-link: facebookresearch/FBGEMM#2070 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: D85604160
EddyLXJ
force-pushed
the
export-D85604160
branch
from
dceec2b to
460f02f
Compare
EddyLXJ
added a commit
to EddyLXJ/torchrec
that referenced
this pull request
Oct 30, 2025
…#3490) Summary: X-link: pytorch/FBGEMM#5062 X-link: facebookresearch/FBGEMM#2070 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. Reviewed By: emlin Differential Revision: D85604160
EddyLXJ
added a commit
to EddyLXJ/FBGEMM-1
that referenced
this pull request
Oct 30, 2025
Summary: X-link: meta-pytorch/torchrec#3490 X-link: facebookresearch/FBGEMM#2070 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. Reviewed By: emlin Differential Revision: D85604160
EddyLXJ
force-pushed
the
export-D85604160
branch
from
460f02f to
785253b
Compare
EddyLXJ
added a commit
to EddyLXJ/torchrec
that referenced
this pull request
Oct 30, 2025
…#3490) Summary: X-link: pytorch/FBGEMM#5062 X-link: facebookresearch/FBGEMM#2070 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. Reviewed By: emlin Differential Revision: D85604160
EddyLXJ
added a commit
to EddyLXJ/FBGEMM-1
that referenced
this pull request
Oct 30, 2025
Summary: X-link: meta-pytorch/torchrec#3490 X-link: facebookresearch/FBGEMM#2070 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. Reviewed By: emlin Differential Revision: D85604160
…#3490) Summary: X-link: pytorch/FBGEMM#5062 X-link: facebookresearch/FBGEMM#2070 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. Reviewed By: emlin Differential Revision: D85604160
EddyLXJ
force-pushed
the
export-D85604160
branch
from
785253b to
7b95615
Compare
EddyLXJ
added a commit
to EddyLXJ/FBGEMM-1
that referenced
this pull request
Oct 31, 2025
Summary: X-link: meta-pytorch/torchrec#3490 X-link: facebookresearch/FBGEMM#2070 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. Reviewed By: emlin Differential Revision: D85604160
meta-codesync bot
pushed a commit
to pytorch/FBGEMM
that referenced
this pull request
Nov 1, 2025
Summary: Pull Request resolved: #5062 X-link: meta-pytorch/torchrec#3490 X-link: https://github.com/facebookresearch/FBGEMM/pull/2070 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. Reviewed By: emlin Differential Revision: D85604160 fbshipit-source-id: 177ec779960a4ac9bfc3d41f38beeb7e56665db8
Bernard-Liu
pushed a commit
to ROCm/FBGEMM
that referenced
this pull request
Nov 11, 2025
Summary: Pull Request resolved: pytorch#5062 X-link: meta-pytorch/torchrec#3490 X-link: https://github.com/facebookresearch/FBGEMM/pull/2070 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. Reviewed By: emlin Differential Revision: D85604160 fbshipit-source-id: 177ec779960a4ac9bfc3d41f38beeb7e56665db8
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.
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: D85604160