1. Proposed title of this feature request

Workload partitioning per node role / MachineConfigPool

2. What is the nature and description of the request?

Simple example, see 3. for another use case:

In a mixed cluster that consists of:

• master nodes
• infra nodes
• worker nodes with PerformanceProfile A
• worker nodes with PerformanceProfile B

We would like to enable Workload Partitioning only on PerformanceProfile B, but not on the masters, nor on the infra nodes nor on PerformanceProfile A.

The current implementation of workload partitioning has a few limitations:
a. As far as I can tell, workload partitioning is currently an all or nothing approach.
b. Workload partitioning can be enabled at install time only, and only for entire clusters.
c. Adding nodes that are not configured for CPUPartitioning is not supported.
d. According to [1], workload partitioning "does not address mixing nodes with pinning and without, this feature will be enabled cluster wide and encapsulate both master and worker pools. If it's not desired then the default behavior will still be turned on but the management workloads will run on the whole CPU set for that desired pool."

[1] https://github.com/openshift/enhancements/blob/master/enhancements/workload-partitioning/wide-availability-workload-partitioning.md

3. Why does the customer need this? (List the business requirements here)

In this specific case, the customer wants to make the master nodes schedulable, add a PerformanceProfile to it and make sure that OpenShift workloads are protected by workload partitioning. However, a preexisting PerformanceProfile on a specific worker node role shall not change its behavior.

4. List any affected packages or components.

Telco