Scaling Twitter for 600 Requests per Second

Add Your Comments

We’ve previously noted the TechCrunch Effect, in which a storm of traffic from bloggers brings new sites to their knees, a la Slashdot. One of 2007′s most popular new services is Twitter, which had its breakout moment during the SXSW Conference in Austin in early March – followed closely be performance problems as the company’s infrastructure struggled to keep pace with its audience.

Twitter was built atop Ruby on Rails, and its scalability challenges provide an interesting case study of managing growth on that platform. Todd Hoff at High Scalability has assembled a good summary of Twitter’s experience, assembled through several presentations by the Twitter team about how they responded to the site’s performance issues. That includes an overview of the process from Blaine Cook, Twitter’s lead architect. Here are the slides from Blaine’s presentation:

There is also a video of the talk that has been posted on Google video, which is about 30 minutes long.

About the Author

Rich Miller is the founder and editor at large of Data Center Knowledge, and has been reporting on the data center sector since 2000. He has tracked the growing impact of high-density computing on the power and cooling of data centers, and the resulting push for improved energy efficiency in these facilities.