Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Address issues with running DRA driver on GKE #23

Merged
merged 5 commits into from
Nov 21, 2023
Merged

Address issues with running DRA driver on GKE #23

merged 5 commits into from
Nov 21, 2023

Conversation

elezar
Copy link
Member

@elezar elezar commented Nov 21, 2023

These changes allow the DRA driver to run on GKE (which supports enabling alpha features).

They include:

  • Scripting to set up a GKE cluster
  • Demos showing device selection based on constraints
  • Updates to the CDI spec generation to handle the GKE node configuration.

Running the included demo with the image ghcr.io/nvidia/k8s-dra-driver:e4a95c14-ubuntu20.04:

 kubectl get nodeallocationstates.nas.gpu.resource.nvidia.com -A -o=json \
        | jq -r '.items[]
             | select(.spec.allocatedClaims)
             | "\(.metadata.name):",
             (.spec.allocatedClaims[])'

gke-k8s-dra-driver-cluster-pool-1-82ea8e4c-02x4:
{
  "claimInfo": {
    "name": "inference-pod-gpu",
    "namespace": "kubecon-demo",
    "uid": "39f74f77-98c4-44bc-9c46-125e5700a60a"
  },
  "gpu": {
    "devices": [
      {
        "uuid": "GPU-679f75dd-b95c-ef89-d691-f3c4f523d43b"
      }
    ]
  }
}
gke-k8s-dra-driver-cluster-pool-2-2afcfde8-86sd:
{
  "claimInfo": {
    "name": "training-pod-gpu",
    "namespace": "kubecon-demo",
    "uid": "5ca7c648-ec6a-4228-ac05-6af75dc36d4c"
  },
  "gpu": {
    "devices": [
      {
        "uuid": "GPU-66e80786-6056-af90-1aaa-10aebff0155a"
      }
    ]
  }
}

and

➜  k8s-dra-driver git:(dra-on-gke) ✗ kubectl logs -n kubecon-demo inference-pod
GPU 0: Tesla T4 (UUID: GPU-679f75dd-b95c-ef89-d691-f3c4f523d43b)
➜  k8s-dra-driver git:(dra-on-gke) ✗ kubectl logs -n kubecon-demo training-pod
GPU 0: Tesla V100-SXM2-16GB (UUID: GPU-66e80786-6056-af90-1aaa-10aebff0155a)

@elezar elezar marked this pull request as ready for review November 21, 2023 10:49
@elezar elezar merged commit 1c4bf38 into main Nov 21, 2023
3 checks passed
@elezar elezar deleted the dra-on-gke branch November 21, 2023 12:26
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants