Skip to main content
Static policies apply explicit request and limit values to matching workloads. Use them when you want deterministic resource values instead of recommendation-driven sizing.

Scope Mapping

  • StaticPolicy is namespaced and references an AutomationStrategy in the same namespace.
  • ClusterStaticPolicy is cluster-scoped and references a ClusterAutomationStrategy.

Field Reference

StaticPolicy.spec

ClusterStaticPolicy.spec

Example: Namespaced Static Policy

Example: Cluster Static Policy

Notes

  • Use static policies when exact values matter more than recommendation-driven tuning.
  • resources.containers."*" applies a default to every scoped container, and named values override wildcard values for the same resource.
  • When scope.containers is set, entries for other containers are removed and wildcard values expand into each listed name.
  • Container targets include regular containers and native sidecar init containers (restartPolicy: Always). One-shot init containers are not automated.
  • Model is included in default workload types. Use spec.scope.workloadTypes: [Model] when you want to restrict scope to KubeAI Model objects only. Recommendations and rollback state are stored on the Model owner, then inherited by model-owned pods.
  • By default, only one same-kind static policy contributes to a workload. With globalConfiguration.multiPolicyContainerRightsizingEnabled: true, separate container/resource targets can come from different policies; PolicyEvaluation resolves overlapping targets. See Multi-Policy Container Rightsizing.