User:Sharihareswara (WMF)/Performance guidelines

From MediaWiki.org
Jump to navigation Jump to search

These performance guidelines aim to help MediaWiki developers avoid common performance problems that slow down MediaWiki and the site. Moved most to Performance guidelines and Performance profiling for Wikimedia code.

Job queue:

The job queue does things with the database. We use Redis to do most of the heavy lifting - to store queue itself. But the runner part is still in-house.

Sources[edit]

Presentations and documents[edit]

  1. 2010 - Roan's Wikimania talk on security, scalability and performance for extension developers
  2. July 2011 - Tim's security & performance talk
  3. August 2011 - Tim on performance
  4. March 2012 - Asher on site performance, graphite, and gdash
  5. May 2012 - "Scalable Web Architecture and Distributed Systems" by Kate Matsudaira
  6. June 2012 - Roan's MySQL optimization tutorial (slides)
  7. September 2012 -
    Asher's Sept 2012 performance profiling tutorial
  8. February 2013 - Sumana on graphite and ganglia
  9. spring 2013 - Tim's performance talk at Amsterdam hackathon
  10. January 2014 - Architecture Summit notes on performance
  11. January 2014 - Graphite docs
  12.  ? - Job class reference
  13. March 2014 - Our use cases for Redis
  14. April 2014 - Manual:Job queue (and, from November 2013, Manual:Job queue/For developers)
  15.  ? - Manual:How_to_debug#Profiling
  16. April 2014 - http://ljungblad.nu/post/83400324746/80-of-end-user-response-time-is-spent-on-the-frontend