Skip to content

[GCS TPU7X NPI Recipe] Update GCS recipe to use GCS Direct and XLA Cache#207

Open
lepan-google wants to merge 1 commit intoAI-Hypercomputer:mainfrom
lepan-google:tpu7x-gcs-update
Open

[GCS TPU7X NPI Recipe] Update GCS recipe to use GCS Direct and XLA Cache#207
lepan-google wants to merge 1 commit intoAI-Hypercomputer:mainfrom
lepan-google:tpu7x-gcs-update

Conversation

@lepan-google
Copy link
Copy Markdown
Contributor

@lepan-google lepan-google commented Mar 28, 2026

As previous model streamer GCS direct and vLLM issues were fixed in the new vLLM image. We update the recipe to use GCS direct and XLA cache as the recommended storage option per the previous investigation link.

TESTED=local tests(server logs, server manifests, client logs, client manifests, manual requests)

@junjieqian
Copy link
Copy Markdown
Collaborator

Hi @lepan-google is this one ready to merge? Can you have someone from storage team to review? I can merge it after. Thank you

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