Performance degradation due to PR #4383 to reduce no-op wakeups in multi-threaded scheduler #4869
Unanswered
manaswini05
asked this question in
Q&A
Replies: 1 comment 1 reply
-
|
This is the first time I hear about that PR causing a performance degradation, but apparently you are not the only one. Can you please open an issue instead of a discussion? Discussions are for general help with Tokio, whereas issues are for bugs. |
Beta Was this translation helpful? Give feedback.
1 reply
-
|
Yeap sure! |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hey! I recently tried upgrading tokio from v1.13.1 to v1.16.1 and observed quite a performance degradation in a real-life service. Reverting this PR - #4383 - seems to fix it. We believe there is reduced concurrency as our CPU usage drops and our response times increase significantly. Were there any rammifications observed with the mentioned PR or any ideas here? Thank you!
Beta Was this translation helpful? Give feedback.
All reactions