sd21-l4-triton-deploy.yaml

apiVersion: v1
kind: Service
metadata:
  name: sd21-l4-triton
spec:
  selector:
    app: sd21-l4-triton
  ports:
    - port: 80
      targetPort: 8000
  type: NodePort
---
apiVersion: apps/v1
kind: Deployment
metadata:
  labels:
    app: sd21-l4-triton
  name: sd21-l4-triton
spec:
  selector:
    matchLabels:
      app: sd21-l4-triton
      albapp: sd21
  template:
    metadata:
      labels:
        app: sd21-l4-triton
        albapp: sd21
    spec:
      nodeSelector:
        karpenter.sh/nodepool: amd-nvidia-l4
      serviceAccountName: appsimulator
      volumes:
      - name: dshm
        emptyDir:
          medium: Memory
      containers:
      - name: app
        image: 891377065549.dkr.ecr.us-west-2.amazonaws.com/stablediffusion:amd64-cuda
        imagePullPolicy: Always
        volumeMounts:
          - mountPath: /dev/shm
            name: dshm
          - mountPath: /app
            name: workdir
        command: ["/run-sd.sh"]
        resources:
          limits:
            nvidia.com/gpu: 1
        ports:
          - containerPort: 8000
            protocol: TCP
        readinessProbe:
          httpGet:
            path: /readiness
            port: 8000
          initialDelaySeconds: 60
          periodSeconds: 10
        env:
        - name: APP
          value: "sd21-l4-triton"
        - name: NODEPOOL
          value: "amd-nvidia-g6"
        - name: NUM_OF_RUNS_INF
          value: "10"
        - name: DEVICE
          value: "triton"
        - name: MODEL_ID
          value: "stabilityai/stable-diffusion-2-1"
        - name: COMPILED_MODEL_ID
          value: "stable-diffusion-2-1"
        - name: POD_NAME
          valueFrom:
            fieldRef:
              fieldPath: metadata.name
      volumes:
      - name: workdir
        emptyDir: {}
      - name: dshm
        emptyDir:
          medium: Memory