Facebook engineer Jason Sobel was with the company just a few weeks in April 2007 when he was asked to help open the company’s new data center in Ashburn, Virginia. In a post today on the Facebook Engineering Blog, Sobel provides some background on the rationales for the data center expansion:
The primary reason for building a new datacenter on the east coast was latency. It takes about 70 milliseconds to send a packet across the country on a high-speed link, and it can be much longer for an average internet user. By putting servers in Virginia we could reduce the time to send a page to users on the east coast and in Europe by a noticeable amount.
A shortage of space and power in the company’s primary data center in Silicon Valley was also a motivation, as was the need for more than one location for disaster recovery purposes. He says the facility in DuPont Fabros’ ACC4 data center now serves “a substantial portion of Facebook’s traffic,”
Sobel also describes the challenges in getting data from the Ashburn and Silicon Valley data centers to integrate properly. “The main scaling challenge with this architecture is pretty obvious: all write operations must happen in one location,” he writes. “Going forward we’re very excited to develop new technologies that will let us perform writes in any location.” Read more at Jason’s entry on the Facebook Engineering blog.