Skip to content

[chart]: cluster-proportional-autoscaler #364

@dylanmtaylor

Description

@dylanmtaylor

Category

Helm Chart

Package/Image/Chart Name

cluster-proportional-autoscaler (chart + underlying image)

Versions

Chart: 1.1.0 (latest). Underlying image: registry.k8s.io/cpa/cluster-proportional-autoscaler:v1.10.0.

Details

cluster-proportional-autoscaler (CPA) is a Kubernetes-SIGs-maintained controller that scales a target deployment proportionally to either the cluster's node count or its core count. It's the standard mechanism for scaling cluster add-ons that need to grow with cluster size — most notably CoreDNS — and is also widely used to scale capacity-reservation pause-pod deployments alongside karpenter / cluster-autoscaler. Both the controller image and the upstream Helm chart are currently absent from the DHI catalog.

The controller is a single small Go binary with a config-map dependency, so a hardened build should port cleanly onto a dhi/golang (build) and dhi/static or dhi/debian-base (runtime) base.

DHI shipping this would unblock CoreDNS-on-DHI deployments at the same time, since CPA is the canonical CoreDNS autoscaler.

Pre-submission Checklist

  • I have searched existing issues to ensure this image, package or Helm chart hasn't been requested before
  • I have provided all the requested information above

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions