Facebook Now Running 10,000 Web Servers

Add Your Comments

How big is Facebook’s Internet infrastructure? Facebook VP of Technology Jeff Rothschild provided some details in a panel at the recent MySQL user conference. Rothschild says Facebook is now running 10,000 servers, including 1,800 MySQL servers that are overseen by just two database administrators.

That server growth is one reason Facebook has expanded its infrastructure through major leases of new data center space from DuPont Fabros (DFT) in Ashburn, Virginia and Digital Realty Trust (DLR) in Santa Clara, Calif. That’s also why IBM is introducing new iDataPlex servers optimized for Web 2.0 and cloud computing apps later today at the Web 2.0 Conference. IBM has clearly taken note of the success of Rackable (RACK) in large server sales to Facebook, Amazon (AMZN), Microsoft and Yahoo.


James Hamilton points to a summary and video of the panel at the recent MySQL user conference on “Scaling MySQL – Up or Out.” IN addition to Rothschild, the panel included particpants from MySQL, Sun Microsystems (JAVA), Yahoo’s Flickr service, Fotolog, Wikipedia and YouTube, who offer up some numbers on the database infrastructure supporting their sites.

Hamilton notes Facebook’s extensive use of Memcached, the object caching system used to speed up database-driven dynamic web applications:

The Facebook fleet has grown fairly dramatically of late. For example, Facebook is the largest Memcached installation and the most recent reports I had come across have 200 Memcached servers at facebook. At the Scaling MySQL panel, they report 805 Memcached servers.

Facebook’s willingness to talk about the scope of its infrastructure offered a contrast to several other panel participants, most notably the panelist from YouTube, who couldn’t say how many MySQL servers Google is running.

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.