The company says its Open Compute Project will give customers more choice and flexibility in choosing servers
Facebook has proposed a new model for designing servers that it says will give businesses more choice in selecting components and a smarter way to upgrade systems when needs change, though it remains to be seen how widely its method will be adopted.
The social networking company is looking to “disaggregate” the data center, meaning it wants to reduce the dependencies among computer parts and make it easier for large companies to design the systems they need to suit their own particular workloads.
It laid out the plans at a summit on Wednesday for its Open Compute Project, in which it’s working with component suppliers and end-user companies to define specifications for these interchangeable parts.
The idea is that tech-savvy companies would be able to choose a server design more in tune with their needs. They could order the server through a systems integrator, which would then source parts through components suppliers that have signed onto the project. However, there’s some debate about how applicable the model is outside Internet companies and cloud service providers.
Systems from top-tier server vendors are based on standards to a degree, but the variety of configurations those vendors offer is limited, and many parts are soldered to the motherboard so they can’t be swapped out easily. One result of Facebook’s project may be to pressure those vendors into making their own designs more flexible.
Momentum behind the effort is growing. Facebook announced several new Open Compute Project members at the summit, including storage vendors EMC, Sandisk and Fusion-io, and ARM processor vendors Calxeda, Applied Micro and Tilera. Hitachi also joined, as did Orange and NTT Data, who joined the roster of companies that use servers and are providing input.
Facebook also announced new specifications that expand the range of systems available through the project and could make them applicable to more customers. For example, it is developing a common processor slot that will allow companies to put CPUs from different vendors in the same motherboard.
Meanwhile, Intel said it will contribute its silicon photonics technology, to provide a fast interconnect within server racks, and Advanced Micro Devices said it completed a motherboard specification for use by financial services companies.
Facebook’s argument is that current server designs are too inflexible. Customers who buy from top-tier vendors such as Dell or Hewlett-Packard have limited choice among the components they can pick, and they often can’t change those parts once they’re in place.
“The power supply shouldn’t be embedded in the server, otherwise you have to design the server around it, and if your power requirements change, you’re stuck,” said Frank Frankovsky, a Facebook vice president, in a speech opening the summit.
Instead, power supplies can be implemented at the rack level, where they can be upgraded as needs dictate, he said.
Likewise, I/O modules, CPUs and other components should not be dependent on each other, Frankovsky said.
“That will allow us to do a smarter technology refresh when it’s time to upgrade, where we take out some components but leave others we don’t need to change,” he said.
The common processor slot will allow companies to “evaluate different CPUs right up to the last hour” when they’re selecting a new server design, he said.
It’s easy to see how the model could benefit large-scale Internet services companies such as Facebook, which already designs its own servers. It has also attracted cloud service providers such as Rackspace, which also submitted new designs this week.
But some participants here said other large companies can benefit, too.
AMD said the motherboard it developed is designed to bring the Open Compute Project to more than just online giants such as Facebook. It developed the specification, which was codenamed Roadrunner, by working with financial services firms Fidelity Investments and Goldman Sachs.
It fits into a standard server rack, something that already sets it apart from Facebook’s own server design, which is for a custom rack. The AMD board includes its Opteron 6200 or 6300 processors, but it allows the customer to choose the I/O module and whether they want a SAS controller, said Bob Ogrey, AMD’s “cloud evangelist” and the engineer who designed the board. It also implements a low-cost server management platform that AMD developed with Broadcom, he said.
Companies are testing the board now, and full production will begin at the end of the quarter, he said.
Matt Eastwood, an analyst with IDC, said it makes sense for companies that rely on IT for a competitive advantage to design their own servers. Big banks are willing to invest heavily in powerful systems to run their Monte Carlo simulation programs, for example, which are used to assess investment risk.
Facebook’s model may also suit other companies for whom high-performance computing is essential, Eastwood said, such as pharmaceutical firms or companies that do oil and gas exploration.
Still, that’s far from most of the world’s big businesses. “You’re probably talking about a few hundreds of companies” that have an incentive to make use of the Open Compute Project’s designs, he said.
He also noted that while Facebook is talking about disaggregation, there’s a trend in the industry in the opposite direction, with companies such as Oracle promoting highly integrated systems as a way to achieve maximum compute performance.
Nathan Brookwood, an analyst with Insight64, was also skeptical that the Open Compute project is applicable outside large Internet and cloud service providers. “I’m not sure enterprises can benefit from it,” he said.
Peter ffoulkes, an analyst with 451 Research, was more optimistic. “I think it will be relevant to any large company that wants to customize their data center to match their workload,” he said.
Even large retailers such as Target and Wal-Mart, which may not appear to be technology-driven companies, do a lot of data analysis to uncover customer buying trends, he noted.
The top-tier server vendors don’t want to be left out of the picture, and some said they’ll support the Open Compute specifications in some products. Dell showed two prototype servers, for example, one with an x86 processor and one with an ARM processor, which were both running a server management platform developed through Facebook’s project.
Jay Parikh, Facebook’s vice president of engineering, said the move to more flexible systems couldn’t come at a better time. Data volumes are so high, he said, that companies such as Facebook need to architect complex, custom storage systems to manage the load.
Facebook users upload more than 350 million new photos a day, he said, consuming more than 7 petabytes of storage capacity each month. Those numbers may sound unique to Facebook, but data volumes are expanding for everyone, and other companies will face similar challenges in the near future, according to Parikh.
“The big data challenges that we face today are definitely going to be your big data challenges tomorrow,” he said.
for more information where Total Soft Tech Solutions Inc. gets its news.
visit link below: