Adding new ID_COUNT eviction trigger condition#4829
Closed
EddyLXJ wants to merge 1 commit into
Closed
Conversation
✅ Deploy Preview for pytorch-fbgemm-docs ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D81151216 |
EddyLXJ
added a commit
to EddyLXJ/FBGEMM-1
that referenced
this pull request
Sep 5, 2025
Summary: X-link: facebookresearch/FBGEMM#1855 Differential Revision: D81151216
EddyLXJ
force-pushed
the
export-D81151216
branch
from
055d29e to
b3c2437
Compare
Contributor
|
This pull request was exported from Phabricator. Differential Revision: D81151216 |
EddyLXJ
added a commit
to EddyLXJ/torchrec
that referenced
this pull request
Sep 5, 2025
Summary: X-link: pytorch/FBGEMM#4829 X-link: facebookresearch/FBGEMM#1855 Differential Revision: D81151216
Summary: X-link: meta-pytorch/torchrec#3357 X-link: facebookresearch/FBGEMM#1855 ID COUNT Trigger is based on total num of id in all kvzch tables. Memory usage is cross all tables but eviction trigger is for tbe, like: if target total memory is 10G tbe1 has table [table0, table1], training_id_eviction_trigger_count is [200, 300] tbe2 has table [table2], training_id_eviction_trigger_count is [500] then the target memory for table0 is 2G, table1 is 3G, table2 is 5G, but the trigger condition is tbe level. If currently total id num in tbe1 is [100, 350], in this case tbe1 will not trigger evict even num in table1 is greater than 300 because total num. 100 + 350 = 450 < 500. If total id num in tbe1 is [100, 450], it will trigger eviction on tbe1 only for table1, because num in table0 100 < 200. Differential Revision: D81151216
EddyLXJ
force-pushed
the
export-D81151216
branch
from
b3c2437 to
bbfb1a2
Compare
Contributor
EddyLXJ
added a commit
to EddyLXJ/torchrec
that referenced
this pull request
Sep 11, 2025
Summary: X-link: pytorch/FBGEMM#4829 X-link: facebookresearch/FBGEMM#1855 ID COUNT Trigger is based on total num of id in all kvzch tables. Memory usage is cross all tables but eviction trigger is for tbe, like: if target total memory is 10G tbe1 has table [table0, table1], training_id_eviction_trigger_count is [200, 300] tbe2 has table [table2], training_id_eviction_trigger_count is [500] then the target memory for table0 is 2G, table1 is 3G, table2 is 5G, but the trigger condition is tbe level. If currently total id num in tbe1 is [100, 350], in this case tbe1 will not trigger evict even num in table1 is greater than 300 because total num. 100 + 350 = 450 < 500. If total id num in tbe1 is [100, 450], it will trigger eviction on tbe1 only for table1, because num in table0 100 < 200. Differential Revision: D81151216
facebook-github-bot
pushed a commit
to meta-pytorch/torchrec
that referenced
this pull request
Sep 12, 2025
Summary: Pull Request resolved: #3357 X-link: pytorch/FBGEMM#4829 X-link: facebookresearch/FBGEMM#1855 ID COUNT Trigger is based on total num of id in all kvzch tables. Memory usage is cross all tables but eviction trigger is for tbe, like: if target total memory is 10G tbe1 has table [table0, table1], training_id_eviction_trigger_count is [200, 300] tbe2 has table [table2], training_id_eviction_trigger_count is [500] then the target memory for table0 is 2G, table1 is 3G, table2 is 5G, but the trigger condition is tbe level. If currently total id num in tbe1 is [100, 350], in this case tbe1 will not trigger evict even num in table1 is greater than 300 because total num. 100 + 350 = 450 < 500. If total id num in tbe1 is [100, 450], it will trigger eviction on tbe1 only for table1, because num in table0 100 < 200. Reviewed By: emlin Differential Revision: D81151216 fbshipit-source-id: 693bd47e950cc63498b4cdfc7691636197c832dc
Contributor
|
This pull request has been merged in b94d1ba. |
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
2 participants
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.
Differential Revision: D81151216