beginnerEnablement ~25 min

Deploy the NVIDIA GPU Operator

Install the GPU Operator with Helm and validate GPU enablement.

Scenario

A fresh cluster with GPU nodes needs the full GPU software stack.

Prerequisites
  • ·Helm installed
  • ·cluster-admin access
  • ·GPU nodes joined
Tasks
0 / 3 tasks
  1. bash
    helm repo add nvidia https://helm.ngc.nvidia.com/nvidia
helm repo update
  2. bash
    helm install --wait gpu-operator nvidia/gpu-operator \
  -n gpu-operator --create-namespace

    Expected · The release deploys and pods roll out.

  3. bash
    kubectl get nodes -o json | jq '.items[].status.allocatable' | grep gpu
Success criteria
  • ✓gpu-operator pods Running
  • ✓Nodes advertise nvidia.com/gpu
  • ✓A test pod schedules
Hints
  • ·On RKE2, set the operator's CONTAINERD_CONFIG to the RKE2 containerd path.
  • ·Pre-installed host drivers? Set driver.enabled=false.