Kubernetes调度与资源管理详解

Kubernetes调度与资源管理详解

1. 资源请求与限制

spec: containers: - name: app resources: requests: memory: "128Mi" cpu: "250m" limits: memory: "256Mi" cpu: "500m"

2. 调度策略

2.1 节点亲和性

spec: affinity: nodeAffinity: requiredDuringSchedulingIgnoredDuringExecution: nodeSelectorTerms: - matchExpressions: - key: disktype operator: In values: - ssd

2.2 Pod亲和性

spec: affinity: podAntiAffinity: requiredDuringSchedulingIgnoredDuringExecution: - labelSelector: matchExpressions: - key: app operator: In values: - nginx topologyKey: kubernetes.io/hostname

3. 污点与容忍

# 节点添加污点 kubectl taint nodes node1 key=value:NoSchedule # Pod添加容忍 spec: tolerations: - key: "key" operator: "Equal" value: "value" effect: "NoSchedule"

4. 总结

Kubernetes调度机制通过资源请求限制、亲和性规则等，实现应用的高效调度。