Open Compute Project: What you need to know

What is the Open Compute Project?

The Open Compute project is nonprofit founded in 2011 with the goal of reimaging the design of server, storage and other data center hardware with the goal of driving scalable computing through sharing of information and technical specifications.

The appetite for mobile content delivered at scale and as quickly as possible is driving the build out of data centers around the world. More data centers means content is physically located closer to the end user and, thereby, accessible more quickly.

But operating a data center is a complicated proposition. Massive energy and cooling needs have to be balanced with providing colocation in a manner that’s cost effective for the data center provider and its client.

The Open Compute Project aims to create collaborative workflows designed to foster innovation and result in value for every stakeholder in the value chain.

“We believe that openly sharing ideas, specifications and other intellectual property is the key to maximizing innovation and reducing operational complexity in the scalable computing space,” the organization notes. “The Open Compute Project Foundation provides a structure in which individuals and organizations can share their intellectual property.”

Guiding principles include a focus on creating the “most efficient servers, storage and data center infrastructure from a useful work per total cost perspective, in order to bring computing to people at the lowest cost and widest distribution,” with an emphasis on “minimizing environmental impact wherever possible.”

“The base designs that emerge from this project should be freely implemented and improved upon by anyone and all,” according to the group.

Who’s in charge?

While the Open Compute Project stresses the democratization of hardware, like any effective operation, it does have standing leaders and a clear organizational structure.

The Open Compute Project Foundation has a eight-member board of directors comprising:

• Jason Taylor, VP of infrastructure for Facebook; he’s also chairman and president of the Open Compute Project Foundation.
• Rocky Bullock, executive officer of JumpStart Partner, and secretary/treasurer of the Open Compute Project.
• Don Duet, managing director of Goldman Sachs.
• Bill Laing, corporate VP of cloud and enterprise at Microsoft.
• Jason Waxman, GM of high density computer for Intel’s Data Center Group.
• Mark Roenigk, COO of Rackspace Hosting.
• Andy Bechtolsheim, founder, chief development officer and chairman of Arista Networks.
• Frank Frankovsky, formerly of Facebook and past president and chairman of the Open Compute Project Foundation.

As for organization, the board of directors provides guidance to a CEO who’s supported by an operations director/community manager, a VP of channel development and an events manager. The board also appoints a 10-member incubation committee, which directs activities in eight areas: certification, data center, hardware management, HPC, networking, open rack, server and storage.

Who’s involved?

Open Compute Project has a four-tiered membership structure. At the top are platinum members Facebook, Goldman Sachs, Hyve Solutions, Intel, Microsoft, Nexius, QCT, Penguin Computing, Rackspace and Stack Velocity.

Gold level members are Agema, Big Switch Networks, Nokia, Hewlett-Packard and SK Telecom. Silver level members are Bank of America, Samsung, Inventec and Wiwynn.

The Open Compute Project community comprises: AMAX, AndRobo Hitech Solutions, ASrock Rack, AT&T, BetaLead Technologies, Cavium, Cloudbase, Dell, Eplexity, Ericsson, Hexatom, INPHI Corporation, IBM, Ixia, MarklTx, Memoright International, Mellanox, Midfind Systems, Nutanix, Open IO, Panasonic, Pattern Matched Technologies, Sunwooda Electronic, Toshiba, Toshiba TRM, Schneider Electric, Excelero Storage, LiteOn Technology, Equinix and INSPUR Systems.

This video shares information about getting involved with the Open Compute Project.

What does it do?

Consider Open Compute Project founder Facebook, which has huge data center needs and as a result, huge attendant costs to ensure users can access selfies and argue politics with strangers. According to Business Insider, Facebook said last year Open Compute Project designs saved the company $2 billion in operational costs.

Facebook recently announced it would build a second data center in Europe, this one in Ireland. This particular data center will operate completely using technology developed through the Open Compute Project. The data center compute and storage market is largely dominated by players like Cisco, Dell and others that sell the hardware. The Open Compute Project has ushered in the advent of white box hardware; that is purpose-built hardware based off of open technical specifications and comparatively cheaply manufactured by ODMs.

Consider recent research from the Dell’Oro Group, which found more than one-third of all servers shipped in North America in the first quarter of 2015 were white box servers. Dell’Oro Group director Sameh Boujelbene said cloud data centers are driving the demand for white box servers.

“While the growth in servers destined for cloud deployment has benefited various vendors, including the top U.S.-based server vendors, it has disproportionately bolstered the performance of white box server vendors,” Boujelbene said. “This is because most of the growth in the cloud data centers during the quarter was driven by the big four (Amazon, Facebook, Google and Microsoft), who are mainly deploying white box servers.”

So, in addition to disrupting the data center hardware market, the Open Compute Project is giving companies the ability to make a better piece of hardware that’s tailored to their needs and cheaper than its name-brand comp set.

What’s the tie in to telecom?

Meet the recently announced Open Compute Project Telco Project, “focused on data center technologies for telecommunications companies.”

There are three claimed objectives: communicating telco technical requirements effectively to the OCP community; strengthening the OCP ecosystem to address the deployment and operational needs of telcos; and brining OCP innovations to telco data center infrastructure for increased cost-savings and agility.

In addition to AT&T and SK Telecom, carriers Deutsche Telekom, EE and Verizon Communications all joined the Open Compute Project along with the announcement of the Telco Project.

“Leaders in telecommunications embracing OCP signifies the start of a new and exciting chapter as we work together to enable better designs, easier adoption, and efficiency gains across the board,” explained Open Compute Project President and Chairman Jason Taylor.

Andre Fuetsch, SVP of architecture and design at AT&T, said Open Compute Project collaboration will facilitate the carrier’s goal of virtualizing 75% of its network functions by 2020.

“To do that, we need to move to a model of sophisticated software running on commodity hardware,” Fuetsch said. “We’re becoming a software and networking company. As a result, our central offices are going to look a lot more like data centers as we evolve our networking infrastructure. The Open Compute Project is innovating rapidly in this area, and we’re thrilled to be collaborating with the community of engineers and developers that are driving the evolution. We look forward to our vendors and other industry players supporting this initiative, as well.”

To find out more about the Open Compute Project, click here.