Monthly Archives: September 2013
General Motors’ decision to bring its datacenters back in-house after decades of outsourcing to its spin-off EDS (now part of HP) is indicative of a sea-change in the computing strategies of large enterprises triggered by the rise of cloud connectivity and smart devices. GM’s two new mega datacenters will each house 10,000 or more low-cost servers, reports GigaOm, which are required to service “the explosion of on-board vehicle electronics coupled with GM’s unique need for reliability” (Pro subscription required for full report).
Facebook is leading a revolution in how enterprise hardware is built. About two and half years ago, it launched the Open Compute Project (OCP) to create “open source” data center hardware. That means hardware vendors like HP, Dell and Cisco, who basically own the $150 billion data center hardware market, no longer control the product designs. Customers like Facebook and Goldman Sachs do. Because customers are the designers, OCP’s hardware projects use fewer materials, cost less and perform better than what traditional vendors typically offer.
When Facebook opened its data centre in Luleå in northern Sweden, the first it had built outside of the US, it raised more than a few eyebrows – and questions. Why Sweden? Why so far away from any major population centre? And isn’t it something of a risk to fill it with bespoke equipment based on the Open Compute Project, rather than conventional, off-the-shelf servers from HP or Dell?
Data centre operators can dramatically cut energy costs and their impact on the environment by doing without air conditioning, according to research by Facebook. According to V3.co.uk, the findings come from the firm’s Open Compute Project, aimed at making the social network’s IT operations as efficient as possible. Facebook said that it uses “100 percent outside air” to cool all of its own data centres, and that other data centre operators are typically over-cooling their facilities when they do not really need to do this.
Many tech companies are implementing green practices in manufacturing and development, but Facebook has an original approach: Using potatoes in its servers to make them more environmentally friendly. Under the Open Compute Project (OCP), Facebook is on a mission to improve the efficiencies of the servers, storage devices, and data centers that are used to power its social networking platform. Any breakthroughs that the company makes are shared with the rest of the OCP community so that they too can improve their own efficiencies and reduce the overall environmental impact of IT on the world.
Data centre operators can dramatically cut energy costs and their impact on the environment by doing without air conditioning, according to Facebook. The findings come from the firm’sOpen Compute Project, aimed at making the social network’s IT operations as efficient as possible. Facebook said that it uses “100 percent outside air” to cool all of its own data centres, and that other data centre operators are typically over-cooling their facilities when they do not really need to do this.
Hyve Solutions, a division of SYNNEX Corporation (SNX), announced that it has contributed its Open Compute Project-Ready server design engineered to fit into a standard 19-inch data center rack to the Open Compute Project. Hyve debuted the Hyve 1500 Series platform last January at the Open Compute Summit. Hyve Solutions 1500 Series is OCP V2 and is composed of 28 2-node 1.5u high servers into a 44u rack.
Open Compute Project Rack Solution to be Among Products Demonstrated by Quanta QCT at Intel IDF Event This Week
Visitors to the Intel Developer Forum event (IDF ’13) this week in San Francisco have a chance to witness demonstrations of a new Open Compute (OCP) rack solution and a recently announced microserver solution from Quanta QCT. The products will be running alongside other compute, storage and switch solutions that the company has designed specifically for the needs of customers in the high-performance computing, cloud-ready datacenter and hyperscale datacenter segments of the market.
By Charlie Manese, Facebook Hardware Design team At Facebook, because of our scale, we require that solutions deployed in our data center be engineered for maximum operational efficiency and serviceability. The data center team works closely with the hardware design team to ensure this. Our designs incorporate features such as front-of-rack serviceability, toolless repair operations, and simplicity.