Twitter still has its ups and downs. But they’re mostly ups due to a dramatic improvement in Twitter’s site performance, which once was an almost weekly discussion of debate in the blogosphere. How did the Twitter team address the service’s reliability problems? “Fixing Twitter” will be the topic of one of the sessions at O’Reilly’s Velocity Conference, which is set for June 22-24 in San Jose, California. John Adams from Twitter’s operations team will discuss how the Twitter platform has stabilized and scaled.
“I want to spend some time focusing on the way that process and metrics allow people to really understand how their website works and being able to interpret those metrics and use those for scaling.” Adams said in an interview with O’Reilly’s James Turner, which offers a preview of his talk. “I want to be able to take the lessons from here and put them in a format where they can be applied to anyone’s site. And I think one of the things that we like to show people at Velocity is that there is a different way of looking at performance problems. We encourage getting people into a metrics-driven culture. And small companies and small web shops may not realize that at first of how important it is to use numbers to make decisions”
Adams’ presentation is one of the many presentations at Velocity in which the web architects and operations teams running the Internet’s largest sites will share their experience and expertise on how to build and maintain Web services at Internet scale.
Here are a few examples:
- The June 23 keynote features Jonathan Heiliger, the Vice President of Technical Operations at Facebook, where he oversees global infrastructure, site architecture and IT. Facebook recently surpassed 200 million users worldwide.
- The June 24 keynote speaker will be Marissa Mayer from Google, who oversees the company’s search products and user interface.
- In Surviving the 2008 Elections, Jeremy Bingham of DailyKos.com discusses the trial-and-error performance refinements that helped the site scale for massive traffic during last year’s elections.
- In 10+ Deploys per Day, John Allspaw and Paul Hammond of the photo-sharing site Flickr will discuss deploying new code and the importance of tight cooperation between the development and operations teams.
- Jeremy Zawodny of CraigsList will speak on MySQL Performance from Day 1, which will examine common mistakes and configuration problems that can affect the scalability of the open source MySQL database.
Data Center Knowledge users can save 15 percent on their registration for Velocity by using code vel09data. The event, which has been expanded to a three-day format for 2009, will be held at the Fairmont in downtown San Jose. Special discounted room rates are available for Velocity conference attendees.