Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[Core] Microbenchmark performance regressions in multi_client_put_calls_Plasma_Store and client__tasks_and_put_batch #26195

Closed
stephanie-wang opened this issue Jun 29, 2022 · 1 comment · Fixed by #26202
Closed

[Core] Microbenchmark performance regressions in multi_client_put_calls_Plasma_Store and client__tasks_and_put_batch #26195

stephanie-wang opened this issue Jun 29, 2022 · 1 comment · Fixed by #26202
Labels
bug Something that is supposed to be working; but isn't P0 Issues that should be fixed in short order release-blocker P0 Issue that blocks the release

Comments

@stephanie-wang
Copy link
Contributor

What happened + What you expected to happen

There is a ~40% throughput drop between 6/9 and 6/10, commits 6f3de2a and d0f975e. These are probably related.

Versions / Dependencies

3.0dev

Reproduction script

ray microbenchmark

Issue Severity

No response
@stephanie-wang stephanie-wang added bug Something that is supposed to be working; but isn't P0 Issues that should be fixed in short order labels Jun 29, 2022
@stephanie-wang stephanie-wang added this to the Core Nightly/CI Regressions milestone Jun 29, 2022
@scv119 scv119 added the release-blocker P0 Issue that blocks the release label Jun 29, 2022
@scv119
Copy link
Contributor

scv119 commented Jun 29, 2022 •
edited
Loading

git rev-list 6f3de2a..d0f975e --oneline
d0f975e [RLlib] Fix broken link replay buffer docs. (#25666)
dcfed61 [Core] fix gRPC handlers' unlimited active calls configuration (#25626)
6b9b1f1 [Deployment Graph] Move files out of pipeline folder (#25630)
2546fbf [Serve] Autoscaling for deployment graph (#25424)
1483c45 use smaller instance for scheduling tests (#25635)
271c7d7 [AIR][Serve] Add support for multi-modal array input (#25609)
7c39aa5 [RLlib] Trainer.training_iteration -> Trainer.training_step; Iterations vs reportings: Clarification of terms. (#25076)
94d6c21 [RLlib] Replay Buffer API documentation. (#24683)
c364592 [RLlib] Fix no gradient clipping happening in QMix. (#25656)
730df43 [RLlib] Issue 25503: Replace torch.range with torch.arange. (#25640)
b3a3519 [RLlib] Added meaningful error for multi-agent failure of SampleCollector in case no agent steps in episode. (#25596)
8af9ef8 [RLlib] Discussion 6432: Automatic train_batch_size calculation fix. (#25621)
67b2eca Fix a few type annotations that may confuse people (#25645)
5586b89 [Serve] Improve logs for new Serve REST API (#25610)
445400d [CI] Print a summary of broken links in LinkCheck (#25634)
2614c24 [AIR] Add predict_pandas implementation (#25534)
88524d8 [air] add CustomStatefulPreprocessor (#25497)

@mwtian mwtian mentioned this issue Jun 29, 2022
Merged
stephanie-wang pushed a commit that referenced this issue Jul 5, 2022
@mwtian
Revert "[Core] fix gRPC handlers' unlimited active calls configuration (
ccabba8 
#25626)" (#26202)

Reverts #25626

Closes #26195.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
bug Something that is supposed to be working; but isn't P0 Issues that should be fixed in short order release-blocker P0 Issue that blocks the release
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Revert "[Core] fix gRPC handlers' unlimited active calls configuration" ray-project/ray
2 participants
@scv119 @stephanie-wang