gcloud-lab/apps/base/customer1/embedding-service/embedding-deployment.yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: embedding-service
  namespace: customer1
  labels:
    app: embedding-service
spec:
  replicas: 1
  selector:
    matchLabels:
      app: embedding-service
  template:
    metadata:
      labels:
        app: embedding-service
    spec:
      containers:
        - name: embedding-service
          image: ghcr.io/sirius0xdev/trading-embedding-service:latest
          ports:
            - containerPort: 8000
              name: http
              protocol: TCP
          env:
            - name: MODEL_NAME
              value: "nomic-embed-text-v1.5"
          resources:
            requests:
              cpu: "500m"
              memory: "2Gi"
            limits:
              cpu: "2000m"
              memory: "4Gi"
          startupProbe:
            httpGet:
              path: /health
              port: 8000
            initialDelaySeconds: 60
            periodSeconds: 10
            failureThreshold: 12
          livenessProbe:
            httpGet:
              path: /health
              port: 8000
            initialDelaySeconds: 180
            periodSeconds: 30
            failureThreshold: 3
          readinessProbe:
            httpGet:
              path: /health
              port: 8000
            initialDelaySeconds: 60
            periodSeconds: 10
            failureThreshold: 3