BU Priority Overview

Often customers familiar with OpenShift self-managed are accustomed to having more granular control of their worker nodes than is currently offered in a managed OpenShift cluster, where the control-plane is entirely managed by Red Hat (SRE).

Cloud-native customers that may not be familiar with OpenShift, but come from vanilla Kubernetes familiarity expect to have conventional Kubernetes-native control of the compute infrastructure just as well.

To meet this requirement for customers, this priority is a focus on features and changes that enable customers with the controls they expect in order to meet their cloud budgets, scaling needs and stability goals with confidence when managing their applications on managed OpenShift clusters.

Goals

• Reconsider "guard-rails" that may be hindering more than helping customers that seek to manage their managed OpenShift cloud infrastructure and cluster compute components.
• satisfy existing RFEs related to compute management with the intention of meeting expectations set by the technology ecosystem
• reduce frequency of RFEs related to compute management with the intention of meeting expectations set by the technology ecosystem
• satisfy existing and reduce frequency of support exceptions related to compute management
• reduce toil for the Product managers, Engineers, SREs and Service Delivery staff overall, as a result of the prior goals.
• enable features that help customers save on infrastructure costs where pragmatic

State of the Business

There is a consistent influx of requests from Red Hat field teams and customers, all with a common theme indicating that they need/want more controls made available to them, related to compute management. There are several engineering efforts due to improve in this area to benefit our customers.

Execution Plans

Lay out the plan for how to execute against the BU Priority to meet the previously stated goals. Optionally include criteria for abandonment.

Continue to gather feedback from customers and field teams to rank all related feature requests and support exceptions. Aggregate feedback and prioritise efforts accordingly. 

Continue to drive forward existing backlog of compute-management related features.