Cloud SQL

Cloud SQL provides managed PostgreSQL, MySQL, or SQL Server with automated backups, patching, replicas, and private IP access. Node.js services typically connect through a pooler because serverless and autoscaled instances open many short-lived connections.

					import pg from 'pg';

const pool = new pg.Pool({
  host: process.env.DB_HOST,
  user: process.env.DB_USER,
  password: process.env.DB_PASSWORD,
  database: 'app',
  max: 10,
  idleTimeoutMillis: 30_000,
});
				

| Concern | Practice | |---------|----------| | Connection limits | Pool at app tier or Cloud SQL Auth Proxy / PgBouncer | | Failover | Understand regional HA versus zonal instances | | Migrations | Run in CI with backward-compatible steps | | Secrets | Secret Manager or mounted secrets, not baked into images |

Test restore drills, index strategy, and maintenance windows before relying on HA marketing slides.

On interviews: connection pooling with Cloud Run; read replicas and replication lag; private IP versus public authorized networks; backup PITR and restore testing.

Common pitfalls: one connection per request without pooling; ORM migrations that lock large tables at deploy; treating replica reads as strongly consistent.

The trade-off is reduced DBA toil versus less tuning freedom, connection ceilings, and vendor-specific failover behavior.

Checklist:

• Pool connections from every Node.js runtime tier.
• Prefer private IP and least-privilege DB users.
• Run reversible migrations with monitoring on deploy.
• Schedule and test backup restores regularly.