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

[tune] Increase the minimum number of allowed pending trials for faster auto-scaleup #43455

Merged
merged 4 commits into from
Feb 28, 2024

Conversation

justinvyu
Copy link
Contributor

@justinvyu justinvyu commented Feb 27, 2024

Why are these changes needed?

This PR bumps up the minimum number of allowed pending trials from 16 to 200. This increases the speed of autoscaling for a Tune job that starts with a small cluster.

Ray Core benchmarks test submitting up to 600+ placement groups at a time. Having too many pending placement groups affects actor scheduling latency, so allowing up to 200 pending PGs is a middle ground.

Related issue number

Mitigation for #43451

This does not fully fix the issue, since it may be better to adjust this value dynamically in response to cluster scale-ups. Currently, the value gets fixed to max(16, int(initial_cluster_cpus * 1.1), where initial_cluster_cpus is just the number of CPUs at the beginning of the job.

Checks

  • I've signed off every commit(by using the -s flag, i.e., git commit -s) in this PR.
  • I've run scripts/format.sh to lint the changes in this PR.
  • I've included any doc changes needed for https://docs.ray.io/en/master/.
    • I've added any new APIs to the API Reference. For example, if I added a
      method in Tune, I've added it in doc/source/tune/api/ under the
      corresponding .rst file.
  • I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failures at https://flakey-tests.ray.io/
  • Testing Strategy
    • Unit tests
    • Release tests
    • This PR is not tested :(

Signed-off-by: Justin Yu <justinvyu@anyscale.com>
Signed-off-by: Justin Yu <justinvyu@anyscale.com>
Signed-off-by: Justin Yu <justinvyu@anyscale.com>
Copy link
Member

@woshiyyya woshiyyya left a comment

Choose a reason for hiding this comment

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

LGTM

@justinvyu justinvyu merged commit eda7d7e into ray-project:master Feb 28, 2024
9 checks passed
@justinvyu justinvyu deleted the bump_tune_min_pending_pgs branch February 28, 2024 02:00
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants