Kubernetes — Core Concepts

Kubernetes schedules containers across nodes, self-heals failed pods, and manages scaling — you declare desired state, it reconciles.

When to use

• More than 3 services requiring auto-scaling, self-healing, and rolling deployments
• When you need declarative infrastructure for containerized workloads

Tradeoffs

• Steep learning curve; ~15+ resource types to understand
• etcd as SPOF if not HA; overkill for single-service apps

Deployment + Service

apiVersion: apps/v1
kind: Deployment
metadata:
  name: api
spec:
  replicas: 3
  selector:
    matchLabels:
      app: api
  template:
    metadata:
      labels:
        app: api
    spec:
      containers:
        - name: api
          image: myapp:1.4.2
          ports:
            - containerPort: 8080
          resources:
            requests: { cpu: "100m", memory: "128Mi" }
            limits:   { cpu: "500m", memory: "256Mi" }
---
apiVersion: v1
kind: Service
metadata:
  name: api
spec:
  selector:
    app: api
  ports:
    - port: 80
      targetPort: 8080

HPA

apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: api-hpa
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: api
  minReplicas: 2
  maxReplicas: 20
  metrics:
    - type: Resource
      resource:
        name: cpu
        target:
          type: Utilization
          averageUtilization: 70
    - type: Resource
      resource:
        name: memory
        target:
          type: Utilization
          averageUtilization: 80

Gotcha: A Pod is ephemeral — it can be killed and rescheduled at any time. Never store state in a Pod without a PersistentVolume. Always use a Deployment, never create Pods directly.