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

Upgrade Ray version for Autopilot; shrink worker resource allocation #299

Merged
merged 1 commit into from
Mar 6, 2024

Conversation

artemvmin
Copy link
Collaborator

@artemvmin artemvmin commented Mar 6, 2024

  • Ray version 2.7.1 appears to fix some Ray Autoscaler bugs
  • Lower resource allocation to better utilize node capacity (recommended 1:1 matching between k8s and Ray nodes):
    • Head: c3-standard-4
    • CPU worker:c3-standard-8
    • GPU worker:g2-standard-8
  • Pretty print Ray job info in Jupyter notebook
  • Fix README wording

@artemvmin artemvmin requested a review from imreddy13 March 6, 2024 09:54
@artemvmin artemvmin force-pushed the ray-autoscaler branch 2 times, most recently from 5ed5e81 to 3f5ebbc Compare March 6, 2024 10:15
@artemvmin artemvmin force-pushed the ray-autoscaler branch 5 times, most recently from d4985d4 to c2bea38 Compare March 6, 2024 20:18
@artemvmin
Copy link
Collaborator Author

/gcbrun

@artemvmin artemvmin merged commit c4a9a7b into GoogleCloudPlatform:main Mar 6, 2024
3 checks passed
@artemvmin artemvmin deleted the ray-autoscaler branch March 7, 2024 00:47
blackzlq pushed a commit to blackzlq/ai-on-gke that referenced this pull request Mar 7, 2024
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.

3 participants