Feature Overview (aka. Goal Summary)  

This feature request is based on the proposal outlined here in the Autoscaler for a ProvisioningRequest API object, addressing specific challenges related to random pod grouping and latency in large scale-ups.

Proposed Solution:

Integrate the Kubernetes CA Provisioning Request CRD into OpenShift autoscaler. This CRD allows users to define scaling requirements for groups of pods, enabling efficient and coordinated scaling for related workloads.Provisioning Request  express the fact that groups of pods is connected and should be treated as one entity.

Key Benefits:

• Single Pod Group Awareness:
  • Ensures related pods are scheduled and scaled together.
  • Prevent partial deployments and potential disruptions.
  • Supports all-or-nothing scaling for workloads
• Reduce Scale-up Latency:
  • Enables faster scaling for large deployments (100+ nodes) by provisioning resources atomically in a single operation.
  • Improves application responsiveness and user experience.
  • Supports diverse scaling needs with various provisioning classes.

Goals (aka. expected user outcomes)

• OpenShift Integration: The Provisioning Request CRD should integrate with OpenShift's existing APIs and tooling.
• User Interface: UX within the OpenShift console for managing and monitoring Provisioning Requests.
• Documentation and Examples: Provide clear documentation and examples for OpenShift users on effectively utilizing the Provisioning Request CRD .
• Testing and Validation
•