Postgresql Scaling

PostgreSQL Scaling Skill

Atomic skill for horizontal and vertical scaling

Overview

Production-ready patterns for partitioning, connection pooling, and high availability.

Prerequisites

• PostgreSQL 16+
• Understanding of replication
• PgBouncer or Patroni experience

Parameters

parameters:
  operation:
    type: string
    required: true
    enum: [partition, pool, replicate, shard]
  partition_type:
    type: string
    enum: [range, list, hash]

Quick Reference

Range Partitioning

CREATE TABLE events (id BIGINT, data JSONB, created_at TIMESTAMPTZ)
PARTITION BY RANGE (created_at);

CREATE TABLE events_2024_q1 PARTITION OF events
    FOR VALUES FROM ('2024-01-01') TO ('2024-04-01');

PgBouncer Config

[pgbouncer]
pool_mode = transaction
default_pool_size = 20
max_client_conn = 1000

Replication Setup

CREATE ROLE replicator WITH REPLICATION LOGIN PASSWORD 'secret';
SELECT pg_create_physical_replication_slot('replica1');

Monitor Lag

SELECT client_addr, pg_size_pretty(pg_wal_lsn_diff(sent_lsn, replay_lsn)) as lag
FROM pg_stat_replication;

Pool Mode Selection

Troubleshooting

Usage

Skill("postgresql-scaling")