1---
2title: Use Connection Pooling for All Applications
3impact: CRITICAL
4impactDescription: Handle 10-100x more concurrent users
5tags: connection-pooling, pgbouncer, performance, scalability
6---
7 
8## Use Connection Pooling for All Applications
9 
10Postgres connections are expensive (1-3MB RAM each). Without pooling, applications exhaust connections under load.
11 
12**Incorrect (new connection per request):**
13 
14```sql
15-- Each request creates a new connection
16-- Application code: db.connect() per request
17-- Result: 500 concurrent users = 500 connections = crashed database
18 
19-- Check current connections
20select count(*) from pg_stat_activity;  -- 487 connections!
21```
22 
23**Correct (connection pooling):**
24 
25```sql
26-- Use a pooler like PgBouncer between app and database
27-- Application connects to pooler, pooler reuses a small pool to Postgres
28 
29-- Configure pool_size based on: (CPU cores * 2) + spindle_count
30-- Example for 4 cores: pool_size = 10
31 
32-- Result: 500 concurrent users share 10 actual connections
33select count(*) from pg_stat_activity;  -- 10 connections
34```
35 
36Pool modes:
37 
38- **Transaction mode**: connection returned after each transaction (best for most apps)
39- **Session mode**: connection held for entire session (needed for prepared statements, temp tables)
40 
41Reference: [Connection Pooling](https://supabase.com/docs/guides/database/connecting-to-postgres#connection-pooler)
42