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.

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

[Bugfix] [SpecDecode] AsyncMetricsCollector: update time since last collection #6578

Merged
merged 2 commits into from
Jul 19, 2024

Conversation

tdoublep
Copy link
Member

@tdoublep tdoublep commented Jul 19, 2024

Fix #6577

IDK if I misunderstood something here, but looks like we basically sync the metrics every time (unless one is in-flight) and don't respect the collect_interval_s after the very first collection.

cc @cadedaniel @njhill

Copy link

👋 Hi! Thank you for contributing to the vLLM project.
Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which consists a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of default ones by unblocking the steps in your fast-check build on Buildkite UI.

Once the PR is approved and ready to go, please make sure to run full CI as it is required to merge (or just use auto-merge).

To run full CI, you can do one of these:

  • Comment /ready on the PR
  • Add ready label to the PR
  • Enable auto-merge.

🚀

Signed-off-by: Thomas Parnell <[email protected]>
@tdoublep
Copy link
Member Author

tdoublep commented Jul 19, 2024

I added a test to catch this, it fails on main.

Pretty confident that this is in fact a bug now.

It effectively defeats the whole purpose of trying to copy the tensors periodically 😄

Fixing this can potentially improve performance.

/ready

@github-actions github-actions bot added the ready ONLY add when PR is ready to merge/full CI is needed label Jul 19, 2024
Copy link
Collaborator

@cadedaniel cadedaniel left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Great catch, thanks for adding a test as well :)

@cadedaniel cadedaniel enabled auto-merge (squash) July 19, 2024 20:53
@cadedaniel
Copy link
Collaborator

Can we get a force merge for the intel test? @simon-mo ty

@simon-mo simon-mo merged commit f0bbfaf into vllm-project:main Jul 19, 2024
83 of 85 checks passed
@simon-mo
Copy link
Collaborator

🫡

@tdoublep tdoublep deleted the spec-decode-metrics-fix branch July 19, 2024 21:12
xjpang pushed a commit to xjpang/vllm that referenced this pull request Jul 24, 2024
gnpinkert pushed a commit to gnpinkert/vllm that referenced this pull request Jul 26, 2024
cduk pushed a commit to cduk/vllm-pascal that referenced this pull request Aug 6, 2024
Alvant pushed a commit to compressa-ai/vllm that referenced this pull request Oct 26, 2024
KuntaiDu pushed a commit to KuntaiDu/vllm that referenced this pull request Nov 20, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
ready ONLY add when PR is ready to merge/full CI is needed
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[Bug]: SpecDecode AsyncMetricsCollector _last_metrics_collect_time is never reset
3 participants