September 17-19, 2012

Westin Michigan Avenue Hotel, Chicago

Log in or Sign Up

Performance Scaling Roadmap

log in to bookmark this presentaton

Novice / Talk
September 17th 9 a.m. – 12:15 p.m.
The path to having a terabyte sized database with thousands of users isn't easy, but plenty of installations have made it there there. Learn about the most common scaling issues as PostgreSQL databases grow.

Abstract

There are plenty of small installations of PostgreSQL that happily run without much planning or maintenance. But as your data and user count grows, the amount of work needed to keep things running will too. Those changes don't have to be a surprise though; there are several common growing pains you can prepare for if you know about them in advance. This tutorial will outline the causes and solutions to the most "popular" scaling problems PostgreSQL installations go through, as they go from megabytes to terabytes and from one user to thousands.

Topics will include:

  • System monitoring
  • Query log analysis
  • Vaccum and table cleanup
  • Cache priming
  • Benefits of partitioning
  • Planning for read-only replicas and pooling