Sheet 06
Backend Microservices Fleet
Service instance sizing, core/memory allocation, throughput capacity, and cluster headroom buffer.
Cluster Headroom1018% — OK
Total Pod Instances
29 Pods
Across 6 servicesTotal Compute Cores
110 vCPUs
Allocated CPUTotal Cluster Memory
244 GB RAM
Working memoryAggregate RPS Capacity
74,500 req/s
Peak demand: 6,667Service Deployment Matrix
Adjust instance counts below to observe live headroom changes| Service Name | Instances | Cores / Pod | RAM / Pod | RPS / Pod | Total RPS Cap | Architecture Role & Notes |
|---|---|---|---|---|---|---|
| API Gateway | 9 | 4 | 8 GB | 3,000 | 27,000 req/s | Rate limiting, JWT auth, TLS, edge routing |
| Order Service | 4 | 8 GB | 2,000 | 12,000 req/s | Outbox pattern, order saga, Kafka producer | |
| Payment Service | 4 | 8 GB | 1,500 | 6,000 req/s | Idempotency keys, PCI-DSS scoped zone | |
| Inventory Service | 4 | 8 GB | 2,500 | 10,000 req/s | High-throughput stock locks, Redis cache | |
| Notification Service | 2 | 4 GB | 5,000 | 15,000 req/s | Async worker, Kafka consumer, Push/SMS | |
| Search Service | 4 | 16 GB | 1,500 | 4,500 req/s | OpenSearch query facade & indexing | |
| Fleet Total | 29 | 110 vCPU | 244 GB | — | 74,500 req/s | 1018% Headroom vs Peak |