ripgraphics/system-design-expert

System Design Expert

Architecture Patterns

| Pattern | Use When | Trade-offs | |---------|----------|------------| | Monolith | Small team, MVP, simple domain | Easy to deploy, hard to scale | | Microservices | Large team, complex domain, scale | Flexible, complex ops | | Serverless | Event-driven, variable load | Pay per use, cold starts |

Scalability Checklist

Horizontal Scaling

• [ ] Stateless application (no local sessions)
• [ ] Load balancer configured
• [ ] Database connection pooling
• [ ] Shared cache (Redis)
• [ ] CDN for static assets

Database Scaling

• [ ] Read replicas for read-heavy
• [ ] Sharding for write-heavy
• [ ] Caching layer (Redis)
• [ ] Indexing optimized

Common Patterns

┌─────────────┐     ┌──────────────┐     ┌─────────────┐
│   Client    │────▶│ Load Balancer│────▶│  App Server │
└─────────────┘     └──────────────┘     └──────┬──────┘
                                                 │
                    ┌────────────────────────────┼────────────────────────────┐
                    │                            │                            │
               ┌────▼────┐                 ┌─────▼─────┐               ┌──────▼──────┐
               │  Cache  │                 │  Database │               │ File Storage│
               │ (Redis) │                 │(PostgreSQL)│               │    (S3)     │
               └─────────┘                 └───────────┘               └─────────────┘

Estimation Guidelines

| Users | Architecture | Database | Cache | |-------|--------------|----------|-------| | < 1K | Monolith | Single DB | Optional | | 1K-10K | Monolith + Cache | DB + Read replica | Redis | | 10K-100K | Microservices | Sharded DB | Redis Cluster | | > 100K | Distributed | Multiple DBs | Distributed Cache |

API Design

# RESTful conventions
GET    /users          # List
GET    /users/:id      # Get one
POST   /users          # Create
PUT    /users/:id      # Update (full)
PATCH  /users/:id      # Update (partial)
DELETE /users/:id      # Delete

# Pagination
GET /users?page=1&limit=20

# Filtering
GET /users?status=active&role=admin

# Response format
{
  "data": [...],
  "meta": { "total": 100, "page": 1 }
}