KEDA

# AWS SQS ScaledObject
apiVersion: keda.sh/v1alpha1
kind: ScaledObject
metadata:
  name: sqs-processor-scaler
  namespace: production
spec:
  scaleTargetRef:
    name: sqs-processor
  minReplicaCount: 0
  maxReplicaCount: 30
  triggers:
    - type: aws-sqs-queue
      metadata:
        queueURL: https://sqs.ap-northeast-2.amazonaws.com/123456789/order-queue
        queueLength: "5"           # 메시지 5개당 Pod 1개
        awsRegion: ap-northeast-2
        scaleOnInFlight: "true"    # 처리 중 메시지도 카운트
      authenticationRef:
        name: aws-credentials

---
# IRSA (IAM Roles for Service Accounts) 인증
apiVersion: keda.sh/v1alpha1
kind: TriggerAuthentication
metadata:
  name: aws-credentials
  namespace: production
spec:
  podIdentity:
    provider: aws-eks              # EKS Pod Identity 사용

---
# ScaledJob: 일회성 작업에 적합
apiVersion: keda.sh/v1alpha1
kind: ScaledJob
metadata:
  name: image-processor-job
  namespace: batch
spec:
  jobTargetRef:
    parallelism: 1
    completions: 1
    backoffLimit: 3
    template:
      spec:
        restartPolicy: Never
        containers:
          - name: processor
            image: myapp/image-processor:v2.0
            resources:
              requests:
                cpu: "1"
                memory: "2Gi"
  pollingInterval: 10
  minReplicaCount: 0
  maxReplicaCount: 100
  successfulJobsHistoryLimit: 5
  failedJobsHistoryLimit: 5
  triggers:
    - type: aws-sqs-queue
      metadata:
        queueURL: https://sqs.ap-northeast-2.amazonaws.com/123456789/image-queue
        queueLength: "1"           # 메시지 1개당 Job 1개
        awsRegion: ap-northeast-2
      authenticationRef:
        name: aws-credentials

---
# 멀티 트리거 예제 (SQS + Cron)
apiVersion: keda.sh/v1alpha1
kind: ScaledObject
metadata:
  name: hybrid-scaler
spec:
  scaleTargetRef:
    name: hybrid-processor
  minReplicaCount: 0
  maxReplicaCount: 20
  triggers:
    # SQS 기반 스케일링
    - type: aws-sqs-queue
      metadata:
        queueURL: https://sqs.ap-northeast-2.amazonaws.com/123456789/tasks
        queueLength: "10"
      authenticationRef:
        name: aws-credentials
    # 업무 시간 최소 유지 (9시~18시)
    - type: cron
      metadata:
        timezone: Asia/Seoul
        start: 0 9 * * 1-5         # 월-금 09:00
        end: 0 18 * * 1-5          # 월-금 18:00
        desiredReplicas: "3"       # 최소 3개 유지

# Prometheus 메트릭 기반 스케일링
apiVersion: keda.sh/v1alpha1
kind: ScaledObject
metadata:
  name: prometheus-scaler
  namespace: production
spec:
  scaleTargetRef:
    name: api-gateway
  minReplicaCount: 2
  maxReplicaCount: 50
  triggers:
    - type: prometheus
      metadata:
        serverAddress: http://prometheus.monitoring.svc:9090
        metricName: http_requests_per_second
        threshold: "100"           # Pod당 100 RPS
        query: |
          sum(rate(http_requests_total{
            namespace="production",
            service="api-gateway"
          }[2m]))
        ignoreNullValues: "false"

---
# HTTP Request 수 기반 (Prometheus + Ingress 메트릭)
apiVersion: keda.sh/v1alpha1
kind: ScaledObject
metadata:
  name: ingress-based-scaler
spec:
  scaleTargetRef:
    name: web-frontend
  minReplicaCount: 3
  maxReplicaCount: 100
  advanced:
    horizontalPodAutoscalerConfig:
      behavior:
        scaleDown:
          stabilizationWindowSeconds: 300
          policies:
            - type: Percent
              value: 10
              periodSeconds: 60
        scaleUp:
          stabilizationWindowSeconds: 0
          policies:
            - type: Pods
              value: 4
              periodSeconds: 15
  triggers:
    - type: prometheus
      metadata:
        serverAddress: http://prometheus.monitoring:9090
        threshold: "50"
        query: |
          sum(rate(nginx_ingress_controller_requests{
            namespace="production",
            ingress="web-frontend"
          }[1m])) /
          count(kube_pod_status_ready{
            namespace="production",
            pod=~"web-frontend.*"
          } == 1)

---
# Redis List 기반 스케일링
apiVersion: keda.sh/v1alpha1
kind: ScaledObject
metadata:
  name: redis-worker-scaler
spec:
  scaleTargetRef:
    name: background-worker
  minReplicaCount: 0
  maxReplicaCount: 20
  triggers:
    - type: redis
      metadata:
        address: redis-master.cache:6379
        listName: job_queue
        listLength: "10"           # 10개 항목당 Pod 1개
        activationListLength: "1"  # 1개 이상이면 스케일업 시작
        databaseIndex: "0"
      authenticationRef:
        name: redis-auth

---
# 스케일링 상태 확인 명령어
# kubectl get scaledobject -A
# kubectl describe scaledobject kafka-consumer-scaler -n production
# kubectl get hpa -n production  # KEDA가 생성한 HPA 확인
# kubectl logs -n keda -l app=keda-operator  # KEDA 로그