Epic Goal

• Enable Guest cluster autoscaling

Why is this important?

• Compute capacity adjust dynamically to workloads.
• Remove manual burden.
• Remove spare capacity.
• Save money.

Scenarios

1. Run an openshiftCluster and nodePool out of the box with cluster autoscaler enabled.
2. Enabling clusterAutoscaler in an existing openshiftCluster

Acceptance Criteria

Dependencies (internal and external)

1. ...

Previous Work (Optional):

1. https://github.com/kubernetes/autoscaler
2. https://github.com/openshift/cluster-autoscaler-operator

Open questions::

1. Do we want to run the cluster autoscaler operator?
2. Do we want to install machineAutoscaler and clusterAutoscaler CRDs
3. Do we allow to change the cluster autoscaler config on day 2?

Done Checklist

• Cluster autoscaling is configurable via openshiftCluster and nodePool.
• The hostedControlPlane controller runs a cluster autoscaler on a target namespace.
• A guest cluster is able to autoscale.