KEP-5007: DRA Device Attach Before Pod Scheduled

[KobayashiD27]

• One-line enhancement description: Some network- or fabric-attached devices need to be attached to a node before a pod using them can be scheduled.

• Issue link: DRA: attach devices to nodes #5007

• Other comments:

[k8s-ci-robot]

Hi @KobayashiD27. Thanks for your PR.

I'm waiting for a kubernetes member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.

Once the patch is verified, the new status will be reflected by the ok-to-test label.

I understand the commands that are listed here.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.

[KobayashiD27]

@pohly
I have drafted the KEP. I would appreciate it if we could discuss it further.

[krmayankk]

I am new to this area of CDI, but would appreciate if we can show how this para is more tightly related to the proposal ? For example does this make it easier to attach requested resource directly to the node where the pod is about to be scheduled or some other example like https://cdi.liqid.com/hubfs/Inference-aaS-WP_FNL.pdf

[KobayashiD27]

In the current implementation, device attachment is executed by NodePrepareResource() after the Pod has been scheduled. However, if the device cannot be attached to the node at this stage, NodePrepareResource() will continue to fail, and the user will see the ContainerCreating state persist (internally, it will keep failing indefinitely). This issue is particularly problematic when fabric devices are shared across multiple Kubernetes clusters, as these clusters may compete for resources, leading to situations where device attachment fails.

Implementing this proposal will help avoid the persistent ContainerCreating state. By streamlining the process of attaching the requested resource directly to the node where the Pod is scheduled, the risk of device attachment failure is reduced. Consequently, this enhances the overall reliability of the system and prevents resource wastage.

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: KobayashiD27
Once this PR has been reviewed and has the lgtm label, please assign ahg-g for approval. For more information see the Code Review Process.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

• keps/sig-scheduling/OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[KobayashiD27]

Hi @pohly,
When you have a moment, could you please take a look at my proposal?

CC: @towca @hase1128
Since this is related to CA and DRA, I would appreciate it if you could also review it when you have the chance.

Thank you!

[pohly]

/ok-to-test

I'm a bit time-constrained right now. Not sure whether I can review for 1.33.

[k8s-ci-robot]

@KobayashiD27: The following test failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
pull-enhancements-verify	84c3d1b	link	true	/test pull-enhancements-verify

Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here.