NCP-AIO Workload Management — All Questions With Answers

apiVersion: v1
kind: Pod
metadata:
  name: gpu-workload
spec:
  containers:
  - name: cuda-container
    image: nvidia/cuda:11.0-base
    resources:
      limits:
        nvidia.com/gpu: 2
  nodeSelector:
    gpu-type: tesla-a100

{
  "policy": "deny",
  "resources": ["gpu-cluster-a"],
  "max_gpu_usage": "10%",
  "priority_level": "low"
}

apiVersion: v1
kind: Pod
metadata:
  name: gpu-workload
spec:
  containers:
  - name: cuda-container
    image: nvcr.io/nvidia/k8s/cuda-sample:nbody
    resources:
      limits:
        nvidia.com/gpu: 1
  nodeSelector:
    nvidia.com/gpu.present: "true"

apiVersion: scheduling.k8s.io/v1
kind: PriorityClass
metadata:
  name: high-priority-gpu
value: 1000000
globalDefault: false
description: "High priority for distributed training"

Error: Failed to create pod: pods "gpu-job" is forbidden: 
failed quota: gpu-quota: must specify limits for nvidia.com/gpu