[DPD] DPD CRD changes with version bump to v3.2#427
Open
rgadagot wants to merge 2 commits into
Open
Conversation
XuanCS
approved these changes
Jun 5, 2026
mbnavali
reviewed
Jun 5, 2026
| type: object | ||
| x-kubernetes-map-type: atomic | ||
| type: object | ||
| pdSpec: |
Collaborator
There was a problem hiding this comment.
Discussed offline, this field name was discussed with the team, PMs and SAs.
mbnavali
approved these changes
Jun 5, 2026
mollyheamazon
approved these changes
Jun 5, 2026
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What's changing and why?
The feature in this release is for Disaggregated Prefill and Decoding (DPD) support to HyperPod Inference, separating the prefill and decode phases onto dedicated GPU pods. The solution extends the existing Inference Operator with role-specific Kubernetes Deployments (prefiller and decoder), GPU-to-GPU KV cache transfer via NIXL over EFA, conditional routing based on input length, and independent autoscaling.
Pins the operator deployment to amd64 Linux nodes via nodeAffinity to prevent CrashLoopBackOff errors when Kubernetes schedules the operator pod on ARM64/Graviton nodes in mixed-architecture clusters (operator images are amd64-only).