Last Updated: 2026-01-22
This document provides an overview of the Spyre Operator for IBM Power. You will learn how to set up and configure the necessary software stack to make the Spyre Accelerators available to AI workloads that run on Red Hat OpenShift.
The Spyre Operator allows users and cluster administrators to configure Spyre in Red Hat OpenShift. It works as a centralized one-stop operator that coordinates many essential components, which can hide the complexity of the Spyre enablement processes in the cluster. The operator ensures that pods are scheduled to nodes with available Spyre devices and manages device allocation at runtime.
The Spyre Operator helps your cluster run AI workloads more efficiently. It improves scheduling, manages resources, works with Spyre operators to optimize spyre use, supports AI pipelines, and integrates with IBM AI services.
IBM Spyre enablement stack for Power IBM PCIe5 x16 AI Accelerator adapter (IBM spyre accelerator card)