Database Scaling: From Zero to Millions

As your application grows, your database strategy must evolve.

Scaling Strategies

Vertical Scaling (Scale Up)

• Add more CPU/RAM
• Simple but limited
• Good for: < 100K users

Horizontal Scaling (Scale Out)

• Add more machines
• Complex but unlimited
• Good for: 100K+ users

Read Scaling

Primary DB (writes) 
    ├── Replica 1 (reads)
    ├── Replica 2 (reads)
    └── Replica 3 (reads)

Write Scaling: Sharding

function getShard(userId: string) {
  const hash = hashFunction(userId);
  return hash % NUM_SHARDS;
}

Common Patterns

Connection Pooling

const pool = new Pool({
  max: 20,
  idleTimeoutMillis: 30000
});

Read Replicas

• Route reads to replicas
• Route writes to primary
• Handle replication lag

When to Scale

1. Response time increasing
2. CPU/Memory > 70%
3. Connection limits hit

Scale before you need to - reactive scaling is expensive!