Kueue

license

Apache-2.0(OSI)

source

public(kubernetes-sigs/kueue is the whole controller)

core-gated

ungated(a 15,902-entry recursive tree carries no path matching enterprise, ee, commercial or proprietary)

Apache-2.0 and governed by a Kubernetes Special Interest Group, so there is no vendor and no paid tier for a core to be withheld from. The 15,902-entry recursive tree carries one governing LICENSE and no enterprise, ee or commercial path.

• https://github.com/kubernetes-sigs/kueue recorded 2026-09-15

  Repository page for kubernetes-sigs/kueue: Apache-2.0 license, public and unarchived, described as 'Kubernetes-native Job Queueing'. Establishes the license and that the source is public.

• https://api.github.com/repos/kubernetes-sigs/kueue/git/trees/main?recursive=1 recorded 2026-09-15

  Full untruncated recursive tree of the default branch, 15,902 entries. Licenses are the root LICENSE and vendored dependencies. No path matches enterprise, ee, commercial or proprietary. A tree lists paths; it is cited for that and for nothing about the vendor's offerings.

• https://kueue.sigs.k8s.io/docs/overview/ recorded 2026-09-16

  Kueue's overview: 'You can install Kueue on top of a vanilla Kubernetes cluster' and 'Kueue does not replace any existing Kubernetes components.' A Kubernetes SIG project with no vendor edition; the installable thing is the whole thing.

2,971 GitHub stars. The project ships as container images and a Helm chart and publishes no package-registry artifact, so no usage-volume channel exists for it and the band rests on stars, which measure attention rather than use. The stars route caps at level 3 for that reason.

• https://github.com/kubernetes-sigs/kueue recorded 2026-09-15

  Repository page showing the star count for kubernetes-sigs/kueue.

Banded on Topology-Aware Scheduling rather than on the queueing alone. Kueue's own documentation states that when the lowest topology level is set to node, TAS 'finds a fixed assignment of pods to nodes and injects a NodeSelector to make sure the pods get scheduled on the selected nodes', and podset-required-topology requires every pod of a set to sit in one domain. That is choosing the machines and reserving the set together, which is band 5 here - the Kubernetes bind call executing the choice does not move the decision elsewhere.

• https://kueue.sigs.k8s.io/docs/concepts/topology_aware_scheduling/ recorded 2026-09-15

  Kueue's Topology-Aware Scheduling concept page: 'When the lowest level of Topology is set to node, TAS finds a fixed assignment of pods to nodes and injects a NodeSelector to make sure the pods get scheduled on the selected nodes', and the podset-required-topology annotation requiring all pods within one topology domain.

• https://kueue.sigs.k8s.io/docs/overview/ recorded 2026-09-15

  Kueue overview: 'Kueue decides when a job should wait, when a job should be admitted to start (as in pods can be created) and when a job should be preempted', with quotas, fair sharing and resource fungibility across flavors.