06 test workload.yaml

apiVersion: apps/v1 kind: Deployment metadata: name: gpu-workload-1g-5gb spec: replicas: 2 selector: matchLabels: app: gpu-workload-1g-5gb template: metadata: labels: app: gpu-workload-1g-5gb app-group: gpu-workload spec: containers: - name: gpu-workload-1g-5gb image: nvidia/cuda:12.4.0-runtime-ubuntu22.04 command: - bash - -c - | nvidia-smi -L; sleep infinity resources: limits: cpu: "100m" memory: "128Mi" nvidia.com/mig-1g.5gb: "1" tolerations: - key: "nvidia.com/gpu" operator: "Exists" effect: "NoSchedule"

apiVersion: apps/v1 kind: Deployment metadata: name: gpu-workload-2g-10gb spec: replicas: 1 selector: matchLabels: app: gpu-workload-2g-10gb template: metadata: labels: app: gpu-workload-2g-10gb app-group: gpu-workload spec: containers: - name: gpu-workload-2g-10gb image: nvidia/cuda:12.4.0-runtime-ubuntu22.04 command: - bash - -c - | nvidia-smi -L; sleep infinity resources: limits: cpu: "100m" memory: "128Mi" nvidia.com/mig-2g.10gb: "1" tolerations: - key: "nvidia.com/gpu" operator: "Exists" effect: "NoSchedule"

apiVersion: apps/v1 kind: Deployment metadata: name: gpu-workload-3g-20gb spec: replicas: 1 selector: matchLabels: app: gpu-workload-3g-20gb template: metadata: labels: app: gpu-workload-3g-20gb app-group: gpu-workload spec: containers: - name: gpu-workload-3g-20gb image: nvidia/cuda:12.4.0-runtime-ubuntu22.04 command: - bash - -c - | nvidia-smi -L; sleep infinity resources: limits: cpu: "100m" memory: "128Mi" nvidia.com/mig-3g.20gb: "1" tolerations: - key: "nvidia.com/gpu" operator: "Exists" effect: "NoSchedule"