Why is database connection pooling important?

A connection involves a TCP handshake, authentication and session setup. Creating one per request adds latency and consumes server memory. A pool keeps idle connections ready and hands them to callers, then returns them when the work finishes.

const pool = new Pool({
  max: 20,
  idleTimeoutMillis: 30000,
  connectionTimeoutMillis: 2000,
});

const client = await pool.connect();

try {
  await client.query("BEGIN");
  await client.query("UPDATE accounts SET balance = balance - 100 WHERE id = 1");
  await client.query("COMMIT");
} catch (error) {
  await client.query("ROLLBACK");
  throw error;
} finally {
  client.release();
}

Sizing and failure modes

• Total connections across all instances must stay below the database limit
• A pool per process multiplies connections when you scale replicas
• An external pooler such as PgBouncer can multiplex many clients onto fewer server connections
• Serverless functions need a pooler or a managed proxy because instances are short-lived
• Long transactions hold a connection and reduce effective capacity
• Set a connection timeout so requests fail fast instead of hanging

Sizing is a trade-off, not a formula. Start from the database's comfortable concurrency and divide by the number of application instances. Monitor pool wait time and active connections; if wait time rises while the database is not saturated, the pool may be too small or connections may be leaked.