Ray Webhook Support for Single-Host, Multi-Slice TPUs #453
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
richardsliu
approved these changes
Mar 26, 2024
|
/gcbrun |
umeshkumhar
added a commit
that referenced
this pull request
Mar 27, 2024
* add rag kuberay and jupyterhub image (#440) * Rollback to previous image (#454) * Ray Webhook Support for Single-Host, Multi-Slice TPUs (#453) * Fix incorrect replicaIndex for single-host, multi replica * Fix single-host, multi-slice deletion logic * Update README & simplify workloads.tfvars for RAG (#445) * RAG marketplace updates (#456) * fix RAG marketplace changes --------- Co-authored-by: Chia-Yi Liang <chiayiliang327@gmail.com> Co-authored-by: zlq <zlq@google.com> Co-authored-by: ryanaoleary <113500783+ryanaoleary@users.noreply.github.com> Co-authored-by: imreddy13 <132504814+imreddy13@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR adds support for single-host, multi-slice TPU worker groups, which require the
TPU_NAMEandTPU_WORKER_IDenv vars. Specifically, this PR changes the webhook to inject areplicaIndexlabel for all Ray worker pods requesting TPUs (previously injected amultiHostReplicalabel only for multi-host). Pods which are deleted and restart will be assigned the sameTPU_WORKER_ID,TPU_NAME, andreplicaIndex. This PR was manually tested by creating single-host, single-slice, multi-host, and multi-slice RayClusters and ensuring that all environment variables were injected correctly, even after multiple Pod deletions.