Kubernetes Cost Optimization: Reduce K8s Spending

# Audit current pod resource usage across a namespace
kubectl top pods -n production --containers

# Get detailed utilization metrics
kubectl describe nodes | grep -A 5 "Allocated resources"

# Use metrics server to see per-pod CPU and memory
kubectl get --raw /apis/metrics.k8s.io/v1beta1/namespaces/production/pods | jq '.items[] | {name: .metadata.name, cpu: .containers[0].usage.cpu, memory: .containers[0].usage.memory}'

# production-web.yaml — optimized resource requests
apiVersion: apps/v1
kind: Deployment
metadata:
  name: web-api
  namespace: production
spec:
  replicas: 3
  template:
    spec:
      containers:
      - name: web-api
        image: dodatech/web-api:1.5
        resources:
          requests:
            cpu: "100m"       # Based on P95 usage of 52m + 50% buffer
            memory: "200Mi"   # Based on P95 usage of 135Mi + 50% buffer
          limits:
            cpu: "500m"       # Burst cap
            memory: "512Mi"   # OOM protection

# Deploy Cluster Autoscaler on EKS
helm repo add autoscaler https://kubernetes.github.io/autoscaler
helm install cluster-autoscaler autoscaler/cluster-autoscaler \
  --namespace kube-system \
  --set autoDiscovery.clusterName=dodatech-eks \
  --set awsRegion=us-east-1 \
  --set extraArgs.scale-down-enabled=true \
  --set extraArgs.scale-down-delay-after-add=10m \
  --set extraArgs.scale-down-unneeded-time=10m

# Deploy Karpenter on EKS (faster, more efficient)
helm repo add karpenter https://charts.karpenter.sh
helm install karpenter karpenter/karpenter \
  --namespace karpenter \
  --create-namespace \
  --set clusterName=dodatech-eks \
  --set clusterEndpoint=$(aws eks describe-cluster --name dodatech-eks --query cluster.endpoint) \
  --set defaultProvisioner.requirements[0].key=karpenter.sh/capacity-type \
  --set defaultProvisioner.requirements[0].op=In \
  --set defaultProvisioner.requirements[0].values=["spot"]

# karpenter-provisioner.yaml
apiVersion: karpenter.sh/v1alpha5
kind: Provisioner
metadata:
  name: default
spec:
  requirements:
    - key: karpenter.sh/capacity-type
      operator: In
      values: ["spot", "on-demand"]
    - key: kubernetes.io/arch
      operator: In
      values: ["amd64"]
    - key: node.kubernetes.io/instance-type
      operator: In
      values:
        - "m5.large"
        - "m5.xlarge"
        - "c5.large"
        - "c5.xlarge"
        - "r5.large"
  limits:
    resources:
      cpu: 1000
  provider:
    subnetSelector:
      Name: "*private*"
    securityGroupSelector:
      Name: "*eks*"
  ttlSecondsAfterEmpty: 30

# Install VPA
git clone https://github.com/kubernetes/autoscaler.git
kubectl apply -k autoscaler/vertical-pod-autoscaler/deploy/manifest/

# Create a VPA for a deployment
kubectl apply -f - <<EOF
apiVersion: autoscaling.k8s.io/v1
kind: VerticalPodAutoscaler
metadata:
  name: web-api-vpa
  namespace: production
spec:
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: web-api
  updatePolicy:
    updateMode: Auto
  resourcePolicy:
    containerPolicies:
    - containerName: '*'
      minAllowed:
        cpu: "50m"
        memory: "100Mi"
      maxAllowed:
        cpu: "2"
        memory: "2Gi"
EOF

# Check VPA recommendations
kubectl describe vpa web-api-vpa -n production

# production-quota.yaml
apiVersion: v1
kind: ResourceQuota
metadata:
  name: production-quota
  namespace: production
spec:
  hard:
    requests.cpu: "20"
    requests.memory: "80Gi"
    limits.cpu: "40"
    limits.memory: "160Gi"
    persistentvolumeclaims: 20
    pods: 100
    
---
apiVersion: v1
kind: LimitRange
metadata:
  name: production-limits
  namespace: production
spec:
  limits:
  - max:
      cpu: "4"
      memory: "8Gi"
    min:
      cpu: "25m"
      memory: "64Mi"
    default:
      cpu: "500m"
      memory: "512Mi"
    defaultRequest:
      cpu: "100m"
      memory: "200Mi"
    type: Container

# Install Kubecost
helm repo add kubecost https://kubecost.github.io/cost-analyzer/
helm install kubecost kubecost/cost-analyzer \
  --namespace kubecost \
  --create-namespace \
  --set kubecostToken="dodatech" \
  --set prometheus.nodeExporter.enabled=false \
  --set global.zoneId="production"

# Expose port and view dashboard
kubectl port-forward --namespace kubecost deployment/kubecost-cost-analyzer 9090

# Query cost allocation via API
kubectl exec -n kubecost deploy/kubecost-cost-analyzer -- \
  curl -s "http://localhost:9001/model/costAllocation?window=1d&aggregate=namespace"

# Get savings report
kubectl exec -n kubecost deploy/kubecost-cost-analyzer -- \
  curl -s "http://localhost:9001/model/savingsScheduling?window=30d"

{
    "namespace": "production",
    "totalCost": 4520.15,
    "resources": {
        "cpu": 2100.30,
        "memory": 980.50,
        "storage": 520.35,
        "network": 320.00
    },
    "savings": {
        "clusterRightSizing": 780.00,
        "spotMigration": 1200.50,
        "idleResources": 340.25
    }
}