Facebook has its data centers down quite literally to a science. There are never more than five types of servers in use at a time, and no infrastructure ever sits dormant waiting for services to launch. Matt Kapko reports for CIO.com on all the ways Facebook is at the cutting edge of data center efficiency.
Facing the Details
Facebook has 1.35 billion monthly users, and 930 million photo uploads, 6 billion Likes, and 12 billion messages are added to the data-fest daily. In order to keep up with the mind-boggling level of heat that must all cook up, data centers were designed to use all cold air from the outside; they are “not chilling air at all.” It travels between servers, mixes in a hot aisle, and goes out the other side of the building.
The giant further benefits from its own open source Open Compute Project, which takes a homogenous approach to computing and allows technicians to more easily optimize systems on the fly. Software has a flexibility that hardware does not, and Jason Taylor, vice president of infrastructure, adds:
“We really believe that the entire industry can benefit from efficiency work that we do, and that we can benefit from the industry feeding back and contributing new ideas and designs,” Taylor says. “Fundamentally our company is going to win or lose based on our product, the cost of our infrastructure, and cost efficiency wins on infrastructure is something we like the entire industry to benefit from.”
The technician-to-server ratio at Facebook falls between 1:15,000 and 1:20,000. Just let that one sink in for a second. Increasingly affordable technology will make it so that it becomes even more manageable to incorporate the increasingly gigantic amounts of data being processed every day.
If you would like to read further about the clever ways in which Facebook operates, you can read the full article here: http://www.cio.com/article/2854720/data-center/how-and-why-facebook-excels-at-data-center-efficiency.html