1. Proposed title of this feature request

Request support for configuring multiple Azure VM instance types within a single ARO machine pool to improve resilience during regional capacity shortages. This enhancement would ensure more reliable autoscaling and reduce operational overhead compared to managing multiple single-SKU machine pools.

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

Customers request the ability to configure multiple Azure VM instance types within a single ARO (Azure Red Hat OpenShift) machine pool, similar to the capability available in Azure Kubernetes Service (AKS) via its mixed-VM-size node pool feature (currently in preview).

In the current ARO architecture, each machine pool is strictly associated with a single Azure VM SKU. This limitation prevents customers from creating heterogeneous worker pools that could dynamically select from multiple instance types depending on availability and capacity within the region.

The requested enhancement is to enable an ARO machine pool to define more than one allowed instance type, allowing the cluster autoscaler to automatically provision whichever instance type has available capacity at deployment time.

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

Customers operating in Azure regions with known or intermittent capacity constraints—particularly for specific VM SKUs—require additional resilience in how ARO scales worker nodes.

Customer needs this for : Improved Availability & Scalability, Operational Resilience, Parity with AKS Capabilities, Simplified Architecture, better Cost Optimization

4. List any affected packages or components.

• ARO Machine API / MachineSets / MachinePools