Feature Overview (aka. Goal Summary)

This feature is about providing workloads within an HCP KubeVirt cluster access to gpu devices. This is an important use case that expands usage of HCP KubeVirt to AL and ML workloads.

Goals (aka. expected user outcomes)

• Users can assign GPUs to HCP KubeVirt worker nodes using NodePool API

Requirements (aka. Acceptance Criteria):

• Expose the ability to assign GPUS to kubevirt NodePools
• Ensure nvidia supports the nvidia gpu operator on hcp kubevirt
• document usage and support of nvidia gpus with hcp kubevirt
• CI environment and tests to verify gpu assignment to kubevirt nodepools functions