Open Compute Project Could Increase Facebook’s and the Whole Tech Industry’s Data Center Efficiency

Today, Facebook announced the Open Compute Project, a collaborative endeavor to design the most efficient computing and the most economical data centers possible. Facebook’s head of technical operations, Jonathan Heiliger, explained that the one and a half year project to redesign its servers and data centers has helped Facebook to make its Prineville, Oregon data center’s servers 38% more efficient and 24% cheaper than the servers it used to buy.

To help share the environmental and cost benefits with other companies, Facebook will make its new server and data center designs and schematics freely available.

We live-blogged the the press event held at Facebook’s Palo Alto headquarters, where CEO Mark Zuckerberg explained how new features like real-time commenting and messaging systems requires more computing capacity, necessitating a more efficient data system.

Facebook has been criticized by Greenpeace for planning on using some coal energy in the Prineville center, despite other efforts to minimize environmental impact. The Open Compute Project could help to improve Facebook’s reputation in the green community.

The data centers and servers necessary to run the site aren’t cheap. In September 2010, a study estimated that Facebook was spending $50 million a year on data centers alone, not counting servers, the $200 million investment in its new Prineville, Oregon center, or the planned $450 million investment in another center in North Carolina. To reduce server strain, in 2010 Facebook switched to the HipHop PHP compiler it designed, reducing CPU usage by 50% and improving performance by 1.8 times.

Now, Facebook has re-imagined the concepts of the server and the data center, building from the ground up to radically increase efficiency. It will use a stripped down server chassis and a redesigned power supply. Its Prineville center will use no air conditioning, and will instead cool servers entirely with natural air flow.

These innovations will help Open Compute Project centers to attain a superior power usage efficiency rating, or ratio of total data center power usage to the power delivered to computing equipment. Facebook’s Prineville center now has a PUE rating of 1.07, compared to the industry standard of 1.5.

At the announcement, Facebook brought together leaders from some of the most data-intensive companies in the tech industry to discuss their plans for the Open Compute Project. Zynga says it is considering implement the insights from the project into the massive cloud computing systems that power its games. Though it hasn’t committed to integrating the changes, efficiency is crucial for Zynga, as it increased its server capacity by 75 times over a recent two year period. Rackspace said the project’s energy efficiencies could reduce its energy costs from $10 million to $6 million a year.

Through the Open Compute Project, Facebook has made its work to increase its own data efficiency scalable. If other companies agree to apply the Open Compute Project’s innovations, the aggregate benefit to the environment should quiet critics like Greenpeace. Even if it doesn’t gain traction with third-parties, the efficiency improvements should help Facebook’s site to continue to run swiftly into the future.

Social Media 101

Mediabistro Course

Social Media 101

Get hands-on social media training for beginners in our online boot camp, Social Media 101! Starting September 4, social media and marketing experts will teach you the best practices to be successful on social. Register before July 31 to get $50 OFF with early bird pricing. Register now!

 

Leave a Reply

3 Responses to “Open Compute Project Could Increase Facebook’s and the Whole Tech Industry’s Data Center Efficiency”

  1. Jonathan Heiliger: From Yahoo’s ISP To Facebook’s Infrastructure To Being A North Bridge VC says:

    [...] the stodgy ways of existing networking companies, he initiated the Open Compute project, which developed a set of open-source server and data center technologies that were 38% more efficient and 24% [...]

  2. Jonathan Heiliger: From Yahoo’s ISP To Facebook’s Infrastructure To Being A North Bridge VC | Global Technology University says:

    [...] the stodgy ways of existing networking companies, he initiated the Open Compute project, which developed a set of open-source server and data center technologies that were 38% more efficient and 24% [...]

  3. Jonathan Heiliger: From Yahoo’s ISP To Facebook’s Infrastructure To Being A North Bridge VC | Rikstuff says:

    [...] the stodgy ways of existing networking companies, he initiated the Open Compute project, which developed a set of open-source server and data center technologies that were 38% more efficient and 24% [...]

Get the latest news in your inbox
interested in advertising with inside facebook?

Social Media Jobs
of the Day

Social Media Specialist

California Academy of Sciences
San Francisco, CA

Community Manager

Advanstar Communications
Iselin, NJ

Social Media Journalist, Travel Website

McMurry/TMG
Washington, DC

Social Media Manager

The National Review
New York, NY

Featured Company

Join leading companies like this one and recruit from the nation's top media job seekers on the Mediabistro Job Board. Every job post comes with our satisfaction guarantee. Learn More
 

Our Sponsors

Also from Inside Network:   AppData - Facebook & iOS Application Stats   PageData - Engagement Data on Facebook Pages   Facebook Marketing Bible   Inside Network Research
 
home | site map | advertising/sponsorships | about | careers | contact us | help courses | browse jobs | freelancers | events | forums | content | member benefits | reprints & permissions terms of use | privacy policy Copyright © 2014 Mediabistro Inc. call (212) 389-2000 or email us