Facebook's Open Compute Project Gaining Momentum After First Year
Facebook's Open Compute Project Gaining Momentum After First Year
MENLO PARK, Calif. A little over a year after Facebook launched its Open Compute Project, organizers of the unusual open-source hardware and software initiative report that it is gaining traction among a large number of companies big and small.
Turns out most enterprises want to save money, power from the walls and staff time. The Open Compute Project (OCP), based on much of the Facebook data center architecture and server design schemes, aims to do precisely that. But to get any project off the ground, it takes old-fashioned selling and recruiting.
"Anybody who's worked in open source will tell you: The easiest part is opening-sourcing things; the hardest part is actually building a thriving community, where there are multiple people contributing to that project," Mike Schroepfer, Facebook's vice president of engineering, told a group of journalists in a whiteboard session on the Facebook campus earlier this week.
Recruiting Effort Is Paying Off
Apparently, the recruiting effort is starting to pay off. Companies are now knocking on Facebook's door, instead of vice versa.
Facebook launched the OCP April 7, 2011. It is an unprecedented attempt to open-source the specifications it employs for its hardware and data center to efficiently power a social network comprising 900 million-plus people. The OCP held its second summit event last month in San Antonio. More than 500 attendees came.
For the Open Compute Project, Facebook itself publishes specs and mechanical designs used to construct its motherboards, power supply, server chassis and server and battery cabinets. GigaOm has hard data points on the specs.
The company is also open-sourcing specs for its data center's electrical and mechanical construction, including technical specs and mechanical CAD files.
A lot has happened in the first year, Frank Frankovsky, Facebook engineer and founding board member of OCP, told the folks at the whiteboard session earlier this week.
Much Has Happened in the First Year
"It's amazing how much can happen in a year," Frankovsky said. "In April 2011, when we open-sourced a set of server and data center designs under the name 'Open Compute Project,' we weren't sure what to expect. It was our hope that we could inspire the industry to be a little more open, a little more innovative and a little more focused on energy efficiency.
"It seems to have worked, although there's still a lot more to do."
Major Developments in Year One
Frankovsky laid out some of the major developments OCP has seen in the first year:
- Dozens of new companies have joined as official members of the project, including HP, AMD, Fidelity, Quanta, Tencent, Salesforce.com, VMware, Canonical, DDN, Vantage, ZT Systems, Avnet, Alibaba, Supermicro, and Cloudscaling. HP, Quanta and Tencent have taken the additional step of joining the OCP Incubation Committee, which reviews proposed projects to determine whether they should receive official OCP support.
- New projects have been proposed to the Incubation Committee, including a Facebook design for a vanity-free storage server (code-named Knox) and highly efficient motherboard designs aimed at the specific needs of financial services companies from AMD and Intel (code-named Roadrunner and Decathlete, respectively).
- OCP has begun mapping out a convergence between Open Rack, the OCP's specification for an open-standard server rack design, and Project Scorpio, a similar spec under development by Tencent and Baidu. We expect to merge the two specs in 2013.
- HP and Dell have announced new, clean-sheet server and storage designs (code-named Project Coyote and Zeus, respectively) that will be compatible with OCP's Open Rack specification.
- VMware has announced that it will certify its vSphere virtualization platform to run on OCP gear, and DDN has announced that it will do the same with its Web Object Scaler (WOS) storage system. Canonical has also announced that they will offer "zero-day" certification on OCP servers, meaning that they will work with the OCP to certify new designs before those designs are released.
OCP has launched an official OCP Solutions Provider program to help enable new opportunities for companies to sell and consume technology based on Open Compute Project designs. Companies currently pursuing Solutions Provider status include Hyve, ZT Systems and Avnet, as well as new business units from Quanta and Wistron (called QCT and Wiwynn, respectively) that have been launched to sell directly to consumers.
The move to what is essentially open-source hardware design is a significant departure from strategies of other companies, such as Google, Twitter and Amazon, which closely guard their data center and hardware specifications to maintain a competitive edge in the cutthroat cloud-computing market.
"We think it's time to demystify the biggest capital expense of an online businessthe infrastructure," Jonathan Heiliger, vice president of technical operations at Facebook, said at the time of OCP's launch.
Facebook broke ground on its first dedicated data center in Prineville, Ore., in January 2010. The data center employs an evaporative cooling system to cool the incoming air, as opposed to traditional chiller systems that require more energy-intensive equipment. With the assistance of chip makers AMD and Intel and server providers HP and Dell, Facebook engineers have spent tens of millions of dollars building custom servers and power supplies in the past year.
Major Energy Efficiencies
These data center solutions delivered a 38 percent increase in energy efficiency at 24 percent lower cost, compared with Facebook's existing facilities, Heiliger said.
Moreover, this technology enabled the data center to earn an initial power usage effectiveness (PUE) ratio of 1.07, pushing it well below the Environmental Protection Agency's standard mark of 1.5.
The Green Grid's PUE is an indicator of data center energy efficiency, and Facebook's Prineville plant ranks as good as it gets for a major data center.
James Hamilton, vice president and distinguished engineer at Amazon Web Services, toured Facebook's Prineville data center last year and came away impressed by what he saw.
"I saw an unusually large number of elegant designs ranging from one of the cleanest mechanical systems I've come across, three phase 480VAC directly to the rack, a low-voltage, direct current, distributed uninterruptable power supply system, all the way through to custom server designs," Hamilton wrote in his blog.