Skip to content

Latest commit

 

History

History
85 lines (57 loc) · 3.19 KB

charter.md

File metadata and controls

85 lines (57 loc) · 3.19 KB
Raw

SIG Autoscaling Charter

This charter adheres to the conventions described in the Kubernetes Charter README and uses the Roles and Organization Management outlined in sig-governance.

Scope

Covers development and maintenance of Kubernetes components for automated scaling in Kubernetes. This includes automated vertical and horizontal pod autoscaling, initial resource estimation, cluster-proportional system component autoscaling, and autoscaling of Kubernetes clusters themselves.

In scope

  • Autoscaling-related API objects, such as the HorizontalPodAutoscaler and VerticalPodAutoscaler

  • Autoscaling-related tools, such as the cluster autoscaler, single-component scaling tools (e.g. pod-nanny), and cluster-proportional scaling tools

  • Ensuring API interfaces (the scale subresource) are available and usable to enable other SIGs to write autoscalable objects, and enable people to interact with those interfaces.

Link to SIG section in sigs.yaml

Code, Binaries and Services

  • Components and utilities that take automated action to scale a component on the cluster (e.g. the horizontal-pod-autoscaler or addon-resizer subproject)

  • Components and utilities that take automated action to scale the cluster itself (e.g. the cluster-autoscaler subproject)

  • Special parts of client-go for interacting with with the scaling interfaces used by the HPA (e.g. the scale-client subproject)

Cross-cutting and Externally Facing Processes

  • Reviewing implementations of the scale subresource to ensure that autoscaling behaves properly

  • Coordinating with SIG Instrumentation to ensure that metrics APIs are suitable for autoscaling on.

  • Coordinating with SIG Scheduling to make sure scheduling decisions can interact well with the cluster autoscaler

  • Coordinating with SIG Cluster Lifecycle on integration between the cluster autoscaler and cluster API

  • Coordinating with SIG Node around Kubelet requirements for vertical scaling of pods

Out of scope

  • Testing general cluster performance at scale (this falls under the purview of SIG Scalability).

  • Owning metrics APIs (this falls under the purview of SIG Instrumentation). SIG Autoscaling should collaborate with SIG Instrumentation to ensure that metrics APIs are suitable for using in autoscaling.

Roles and Organization Management

This sig follows adheres to the Roles and Organization Management outlined in sig-governance and opts-in to updates and modifications to sig-governance.

Deviations from sig-governance

  • SIG Autoscaling does not have chairs as a separate entity from tech leads. The tech leads have the responsibility of chairs.

Subproject Creation

SIG Technical Leads