Strong Shortner

Rationale:

Flask: Lightweight, easy to understand, well-documented Gunicorn: Production-ready, handles multiple workers, battle-tested Alternative considered: FastAPI - faster but adds complexity Trade-off: Simplicity over raw performance

Why PostgreSQL? Decision: Use PostgreSQL as primary database

Rationale:

ACID compliance for data integrity Excellent Peewee ORM support Reliability and data durability Alternative considered: SQLite (not suitable for concurrent access) Why Redis? Decision: Add Redis for caching layer

Rationale:

Fast in-memory storage for frequently accessed data Session storage capability Rate limiting support Alternative considered: Memcached (Redis has more features) Status: Infrastructure ready, caching implementation pending Gold tier

Why Nginx as Load Balancer? Decision: Use Nginx for load balancing

Rationale:

Industry standard reverse proxy Excellent performance and stability Built-in health checks Rate limiting capabilities Alternative considered: HAProxy (Nginx has better ecosystem) Why Docker Compose? Decision: Use Docker Compose for orchestration

Rationale:

Simple declarative configuration Easy local development Portable across environments Alternative considered: Kubernetes (overkill for this scale) Why Prometheus + Grafana? Decision: Use Prometheus for metrics, Grafana for visualization

Rationale:

Open source, widely adopted Excellent Flask integration Powerful query language (PromQL) Rich visualization capabilities Alternative considered: Datadog (paid, vendor lock-in)

Built With

• docker
• flask
• redis
• vultr