How karchevskii/disributed_systems works
Project type and purpose
• This is a cloud-native, microservice-based full-stack application (“Distributed Tictactoe”) with:
• A Vue.js SPA frontend
• Three backend FastAPI services: Users-Service, Game-Service, Game History-Service
• Supporting infrastructure: Redis cluster, two PostgreSQL clusters
• Deployed via Kubernetes with Istio service mesh
File structure
• Top-level folders:
• frontend/ – Vue.js SPA, Dockerfile, Nginx, WebSocket support
• users/ – FastAPI user management, OAuth2, PostgreSQL, migrations
• game/ – FastAPI game logic, WebSockets, Redis, event stream
• game-history/ – FastAPI consumer of Redis Streams, PostgreSQL
• k8s/ – Kubernetes manifests, Istio setup, secrets, ingress, services, deployments
• Patterns: each microservice is containerized, has its own Dockerfile, config, and dependency manifest (requirements.txt).
• Observability, Service Mesh and CI/CD at top level: .github/workflows/ci.yaml
README for architecture insights
• Confirms microservice paradigm, real-time via WebSockets, event-driven with Redis Streams
• Istio Ingress Gateway, VirtualServices, Envoy sidecars
• Data stores: Redis for state and streams; CloudNativePG PostgreSQL clusters for persistence
• Observability: Prometheus, Grafana, Kiali, Jaeger