Skip to content

Increase TPU CI node count from 2 to 32#8857

Merged
tengyifei merged 1 commit intomasterfrom
tengyifei-patch-8
Mar 20, 2025
Merged

Increase TPU CI node count from 2 to 32#8857
tengyifei merged 1 commit intomasterfrom
tengyifei-patch-8

Conversation

@tengyifei
Copy link
Copy Markdown
Collaborator

@tengyifei tengyifei commented Mar 19, 2025

Increase our TPU CI capacity by 16 times while also disabling the auto scale-down which makes TPU nodes slow to start.

This should 16x our testing throughput and reduce TPU related PR blocking
@tengyifei tengyifei changed the title Expose min_tpu_nodes Increase TPU CI node count from 2 to 32 Mar 19, 2025
@tengyifei tengyifei marked this pull request as ready for review March 19, 2025 20:16
@zpcore
Copy link
Copy Markdown
Member

zpcore commented Mar 19, 2025

Can you also paste the output of terraform plan in the PR? Just to double check what will be changed. Thanks

@tengyifei
Copy link
Copy Markdown
Collaborator Author

Spot checked the terraform plan internally.

@tengyifei tengyifei merged commit 6a93dcf into master Mar 20, 2025
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