Kubernetes — LLM 服务编排

LLM API 服务部署

# deployment.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: llm-api
  namespace: finance-ai
spec:
  replicas: 3
  selector:
    matchLabels:
      app: llm-api
  template:
    metadata:
      labels:
        app: llm-api
    spec:
      containers:
      - name: llm-api
        image: your-registry/llm-api:v1.0.0
        ports:
        - containerPort: 8000
        env:
        - name: DASHSCOPE_API_KEY
          valueFrom:
            secretKeyRef:
              name: llm-secrets
              key: dashscope-api-key
        - name: REDIS_URL
          value: "redis://redis-service:6379"
        resources:
          requests:
            memory: "512Mi"
            cpu: "250m"
          limits:
            memory: "2Gi"
            cpu: "1000m"
        livenessProbe:
          httpGet:
            path: /health
            port: 8000
          initialDelaySeconds: 30
          periodSeconds: 10
        readinessProbe:
          httpGet:
            path: /health
            port: 8000
          initialDelaySeconds: 5
          periodSeconds: 5
---
apiVersion: v1
kind: Service
metadata:
  name: llm-api-service
  namespace: finance-ai
spec:
  selector:
    app: llm-api
  ports:
  - port: 80
    targetPort: 8000
  type: ClusterIP

Secrets 管理

# 创建 Secret
kubectl create secret generic llm-secrets \
  --from-literal=dashscope-api-key=sk-xxx \
  --from-literal=redis-password=your-password \
  -n finance-ai

HPA 自动扩缩容

# hpa.yaml
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: llm-api-hpa
  namespace: finance-ai
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: llm-api
  minReplicas: 2
  maxReplicas: 10
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 70
  - type: Resource
    resource:
      name: memory
      target:
        type: Utilization
        averageUtilization: 80

Ingress 配置

# ingress.yaml
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: llm-api-ingress
  namespace: finance-ai
  annotations:
    nginx.ingress.kubernetes.io/proxy-read-timeout: "300"
    nginx.ingress.kubernetes.io/proxy-buffering: "off"  # SSE 流式
spec:
  rules:
  - host: api.finance-ai.com
    http:
      paths:
      - path: /
        pathType: Prefix
        backend:
          service:
            name: llm-api-service
            port:
              number: 80

常用命令

# 部署
kubectl apply -f deployment.yaml -n finance-ai

# 查看状态
kubectl get pods -n finance-ai
kubectl logs -f deployment/llm-api -n finance-ai

# 滚动更新
kubectl set image deployment/llm-api llm-api=your-registry/llm-api:v1.1.0 -n finance-ai

# 扩容
kubectl scale deployment llm-api --replicas=5 -n finance-ai