machine_learning/ml_infrastructure/inference-server-performance/server/trtis_deploy.yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: inference-server
  labels:
    name: inference-server
spec:
  replicas: 1
  selector:
    matchLabels:
      app: inference-server
  template:
    metadata:
      labels:
        app: inference-server
    spec:
      dnsPolicy: ClusterFirst
      imagePullSecrets:
      - name: ngc
      priority: 0
      restartPolicy: Always
      schedulerName: default-scheduler
      securityContext: {}
      serviceAccount: default
      serviceAccountName: default
      terminationGracePeriodSeconds: 30
      containers:
      - args:
        - trtserver
        - --model-store=gs://YOUR-BUCKET-NAME/resnet/
        image: nvcr.io/nvidia/tensorrtserver:19.05-py3
        imagePullPolicy: IfNotPresent
        livenessProbe:
          failureThreshold: 3
          httpGet:
            path: /api/health/live
            port: 8000
            scheme: HTTP
          initialDelaySeconds: 5
          periodSeconds: 5
          successThreshold: 1
          timeoutSeconds: 1
        name: inference-server
        ports:
        - containerPort: 8000
          protocol: TCP
        - containerPort: 8001
          protocol: TCP
        - containerPort: 8002
          protocol: TCP
        readinessProbe:
          failureThreshold: 3
          httpGet:
            path: /api/health/ready
            port: 8000
            scheme: HTTP
          initialDelaySeconds: 5
          periodSeconds: 5
          successThreshold: 1
          timeoutSeconds: 1
        resources:
          limits:
            nvidia.com/gpu: "1"
          requests:
            cpu: 1000m
            nvidia.com/gpu: "1"
        securityContext:
          procMount: Default
          runAsUser: 1000
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File