Optimizing PostgreSQL Application Performance with Solaris Dynamic Tracing 
by Robert Lor
June 2009
This Sun BluePrints™ article describes how Solaris DTrace functionality can help with PostgreSQL database application tuning. Specifically, the article includes examples showing specific aspects of PostgreSQL database server operation that can be observed using DTrace probes. The article assumes that the reader has a basic understanding of the Solaris OS, Solaris DTrace, the PostgreSQL database server, and general database application development. It addresses the following topics:
• "The Need for Postgresql Database Tuning" discusses the importance of evaluating database performance and introduces Solaris DTrace.
• "Identifying and Fixing Performance Problems" provides detailed examples of using Solaris Dtrace probes to help analyze various aspects of PostgreSQL database application performance.
Contents
- Introduction
- The Need for Postgresql Database Tuning
- Advantages of Solaris Dynamic Tracing
- Identifying and Fixing Performance Problems
- Analyzing and Fixing Slow Queries
- Optimizing the Use of PostgreSQL Buffer Cache
- Checkpoint Tuning
- Summary
- About the Author
- References
- Ordering Sun Documents
- Accessing Sun Documentation Online
Robert Lor is a senior engineer in Sun's ISV Engineering organization. He has worked on operating system and Java™ 2 Platform, Enterprise Edition (J2EE™) application development and performance optimization for more than 15 years. Recently, he has worked closely with the PostgreSQL community to improve the observability of PostgreSQL by introducing DTrace probes to the database server.
| Rate this blueprint (Log In to vote.) | |
|---|---|
| Choices | Your Vote |
Great |
|
Good |
|
Fair |
|
Poor |
|
Sign up or Log in to add a comment or watch this page.