HPA

v1.16 后允许 minReplicas=0
https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.27/#horizontalpodautoscaler-v2-autoscaling
metrics
- type - ContainerResource, External, Object, Pods, Resource
- containerResource
  - 可以选择目标 container
- resource
target
- type - Utilization, Value, AverageValue
- averageUtilization - 所有 Pod - 只能 Resource
- averageValue
behavior - scaleDown, scaleUp
- policies
  - type: Pods
  - value - 每次变化的 Pod 数量
  - periodSeconds - 间隔
    - 0-1800
- selectPolicy=Max
- stabilizationWindowSeconds
  - 0-3600
  - up 默认 0
  - down 默认 300 - 5 分钟

apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: server
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: server
  minReplicas: 2
  maxReplicas: 10
  metrics:
  - type: Resource # Resource、Pods、Object、External
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 50

desiredReplicas = ceil[currentReplicas * ( currentMetricValue / desiredMetricValue )]

HPA​

HPA