Advanced use of pg_stat_statements: Filtering, Regression Testing & More

Check out http://www.pgconf.us/2015/event/76/ for the full talk details. This talk will go into lessons learned when using the pg_stat_statements extension for collecting query statistics, and present new open-source tools we’ve developed for working with the extension output: pg_query: A Ruby extension to parse SQL queries into parse trees using the PostgreSQL parser. This can be used for filtering pg_stat_statements output (e.g. finding all queries on a certain table) and works with normalized queries, thanks to parser patches. pg_simulator: Tool for loading production statistics into a sandbox database and running EXPLAIN against collected queries. Useful for regression testing before deploying new versions of client applications to production (and finding missing indices, etc). This talk is primarily aimed at database admins trying to monitor database performance, and hackers interested in the tooling & pg_stat_statements bug fixes we’ll propose for the 9.5 cycle. About the Speaker Founder & Tech at pganalyze.com, the Hosted PostgreSQL Performance Monitoring Tool. Helping out ProductHunt.com with their scaling efforts. Author of pg_query (Ruby extension to parse queries using the raw_parser) and pg_simulator (Helper tool for copying production statistics to a test system). I love working with PostgreSQL statistics and visualizing them.
Length: 43:37
Views 563 Likes: 3
Recorded on 2015-03-25 at PGConf US
Look for other videos at PGConf US.
Tweet this video