Skip to content

Add Initial implementation of KEP-77: Elastic Jobs via WorkloadSlices #5528

@ichekrygin

Description

@ichekrygin
Contributor

What would you like to be added:

Initial implementation of KEP-77: Dynamically Sized Jobs via WorkloadSlices

Why is this needed:

This issue tracks the initial implementation of KEP-77, which introduces support for dynamically resizing jobs in Kueue through the concept of WorkloadSlices.

Completion requirements:

  • Implement creation of WorkloadSlice objects upon scaling events (e.g., parallelism change for batchv1/Job).
    Add pod scheduling gates to defer scheduling until slice admission.
    Implement admission logic for WorkloadSlices.
    Support deactivation and garbage collection of preempted slices.
    Update quota and PodSetAssignments to reflect slice scaling.
    Annotate relationships between new slices and the slices they preempt.
    Add feature gate (WorkloadSlices) and opt-in annotation mechanism.
    Ensure compatibility with existing batch/v1.Job semantics.
    Unit and integration test coverage for scale-up and scale-down flows.

This enhancement requires the following artifacts:

  • Design doc
    API change
    Docs update

The artifacts should be linked in subsequent comments.

Activity

added
kind/featureCategorizes issue or PR as related to a new feature.
on Jun 5, 2025
mimowo

mimowo commented on Jul 8, 2025

@mimowo
Contributor

/retitle Add Initial implementation of KEP-77: Elastic Jobs via WorkloadSlices

changed the title [-]Add Initial implementation of KEP-77: Dynamically Sized Jobs via WorkloadSlices[/-] [+]Add Initial implementation of KEP-77: Elastic Jobs via WorkloadSlices[/+] on Jul 8, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    kind/featureCategorizes issue or PR as related to a new feature.

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

      Development

      Participants

      @ichekrygin@mimowo

      Issue actions

        Add Initial implementation of KEP-77: Elastic Jobs via WorkloadSlices · Issue #5528 · kubernetes-sigs/kueue