feat: set GPU extended_resources for Ray head and worker groups#1126
Open
pingsutw wants to merge 3 commits into
Open
feat: set GPU extended_resources for Ray head and worker groups#1126pingsutw wants to merge 3 commits into
pingsutw wants to merge 3 commits into
Conversation
Populate extended_resources on Ray HeadGroupSpec and WorkerGroupSpec from their requests so GPU accelerator (and shared memory) configuration is propagated to KubeRay. Export get_proto_extended_resources from flyte.extend and add unit tests covering head/worker accelerator propagation. Signed-off-by: Kevin Su <pingsutw@apache.org>
Signed-off-by: Kevin Su <pingsutw@apache.org>
dansola
previously approved these changes
May 30, 2026
Signed-off-by: Daniel Sola <daniel@union.ai> Signed-off-by: Daniel Sola <daniel.sola@union.ai>
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
extended_resourceson the RayHeadGroupSpecand eachWorkerGroupSpecfrom theirrequests, so GPU accelerator (device/partition/device-class) and shared-memory configuration is propagated to KubeRay.get_proto_extended_resourcesfromflyte.extendso the Ray plugin (and other plugins) can build the protoExtendedResources.plugins/ray/tests/) covering accelerator propagation.Test plan
uv run --project plugins/ray pytest plugins/ray/tests/test_task.py -vResources(gpu="T4:1")→head_group_spec.extended_resources.gpu_accelerator.device == "nvidia-tesla-t4".Resources(gpu="A100:2")→worker_group_spec[0].extended_resources.gpu_accelerator.device == "nvidia-tesla-a100".extended_resourcesunset on both head and worker specs.