Skip to main content

​
Horizontal Scaling

Scale ingestion: 
kubectl scale deployment lumina-ingestion   --replicas=10   --namespace lumina
Scale workers: 
kubectl scale deployment lumina-worker   --replicas=20   --namespace lumina

​
Autoscaling

Configure HPA: 
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
  name: lumina-ingestion-hpa
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: lumina-ingestion
  minReplicas: 3
  maxReplicas: 20
  metrics:
    - type: Resource
      resource:
        name: cpu
        target:
          type: Utilization
          averageUtilization: 70

​
Database Scaling

Vertical scaling: Increase PostgreSQL resources: 
postgresql:
  resources:
    requests:
      memory: "8Gi"
      cpu: "4"
Read replicas: Add PostgreSQL replicas for read queries.

​
Capacity Planning

Ingestion throughput:
  • 1 pod: ~10K traces/sec
  • 10 pods: ~100K traces/sec
Worker throughput:
  • 1 worker: ~1K traces/sec processing
  • 20 workers: ~20K traces/sec
Storage:
  • 1M traces: ~10GB
  • 10M traces: ~100GB