Key components

The Spyre operator manages the full lifecycle of Spyre related components, which includes:

Device plug-in for Spyre

Exposes the physical Spyre cards as a resource that Kubernetes can schedule and manage.

Automatic node labeling

Labels the nodes automatically within the Red Hat OpenShift cluster that have Spyre Accelerators installed, simplifying workload placement.

Spyre scheduler

Provides scheduling logic to allocate Spyre devices (PF/VF modes) and ensures that workloads are placed on suitable compute nodes.

Validation webhook

Ensures that user workloads and cluster policies are valid. Blocks faulty deployments or invalid CRs. For example, multiple Spyre requests and unsupported configurations.

Health Checker

The Spyre Health Checker is used to monitor the status of Spyre cards and update the SpyreNodeState with their current health information.