GPU News: Cray Awarded $32 Million Contract

1 comment

Surrounding the GPU Technical Conference this week in San Jose – Cray, Cirrascale and Penguin Computing have GPU announcements:

Cray awarded $32 million Swiss contract.  Cray announced that it has signed a contract with the Swiss National Supercomputing Centre (CSCS) to upgrade and expand its Cray XC30 supercomputer. When the upgrade is completed, the system nicknamed “Piz Daint” will be the first petascale supercomputer in Switzerland. Currently a 750 teraflop Cray XC30, it will be transformed to include NVIDIA K20X GPU accelerators. CSCS is the first customer to order a Cray XC supercomputer with NVIDIA GPUs. ”Piz Daint will help advance the research projects of our diverse user community by leaps and bounds,” said Thomas Schulthess, director of CSCS. “With GPU acceleration integrated into Cray’s latest generation supercomputer, the application performance and the energy efficiency of our simulations will improve significantly. We are very excited about the collaborative development of a truly general-purpose, hybrid multi-core system with Cray.” The contract is valued at more than $32 million, and the upgraded system is expected to be operational in 2014.

Cirrascale launches GB5400 GPGPU blade server.  Cirrascale  announced the next generation of its GB5400 blade server supporting up to eight GPU cards. Utilizing a pair of the company’s proprietary 80-lane PCIe switch-enabled risers, the GB5400 supports up to eight discrete GPGPU (General-Purpose Graphics Processing Unit) cards in a single blade. ”We have redesigned the GB5400 to handle the latest cards from leading GPU providers, including NVIDIA and their wide assortment of high-end GPU cards,” said David Driggers, CEO, Cirrascale Corporation. “Our customers and licensed partners in cloud and High Performance Computing are asking for this increased density and performance, while maintaining the ability to scale the solutions they choose. We’re confident that the GB5400 meets these needs, and in fact, surpasses them.” The Cirrascale GB5400 blade server, including the entire GB series line of GPGPU solutions, as well as the Cirrascale proprietary PCIe switch-enabled riser, are immediately available to order and are shipping to customers now.

Penguin Computing unveils Relion 2808GT. Penguin Computing announced the availability of the Relion 2808GT. The Relion 2808GT supports eight GPUs or coprocessors in two rack units and provides a higher compute density that any other server on the market. The GPUs are supported by a dual socket platform based on Intel’s Xeon E5-2600 product family. The Relion 2808GT also features an on board Dual 10GbE BASE-T controller and up to 512GB of ECC memory. “Penguin has been delivering integrated GPU computing clusters since the version 1.0 of this technology,” said CEO Charles Wuischpard. “The new Relion 2808GT platform in conjunction with the latest GPU and coprocessor technology delivers unprecedented levels of performance. The Relion 2808GT enables our HPC customers to further accelerate their research by shortening the time to result for their simulations.”

About the Author

John Rath is a veteran IT professional and regular contributor at Data Center Knowledge. He has served many roles in the data center, including support, system administration, web development and facility management.

Add Your Comments

  • (will not be published)

One Comment

  1. So nice to see GPGPU taking off so quickly.