Cray Deploys Liquid-Cooled CS300 Supercomputer
October 3rd, 2013 By: John Rath
Cray announced that it has won a contract with Mississippi State University to provide a Cray CS300-LC system. The CS300-LC is a liquid-cooled version of the CS300 cluster supercomputer. Expected the be delivered later this year, the 322 teraflop supercomputer will include Intel Xeon E5-2600 v2 processors and Intel Xeon Phi coprocessors. Its a liquid-cooled design that uses warm water heat exchangers instead of chillers to directly cool the compute processors and memory, allowing for a more efficient removal of system heat.
The new system will be located at the University’s High Performance Computing Collaboratory (HPC2), and will serve as the primary performance computing system for shared research. HPC2 supports a coalition of member centers and groups, and the effort spans a wide range of application disciplines, and several of the member centers have extensive physical and experimental modeling and analysis capabilities to complement their computational efforts.
“Our mission is to serve the University, State and Nation through excellence in computational science and engineering, and we are pleased to have the resources of a Cray supercomputer to support our efforts,” said Trey Breckenridge, director of high performance computing at Mississippi State. “With the Cray CS300 system and its advanced liquid-cooling architecture, we will provide our vast user community with a cost-effective and energy-efficient high performance computing system that is also a powerful and technologically-advanced tool for scientific research.”
Cray adds GPU and Phi to XC30
Cray also announced that it has broadened its support for accelerators and coprocessors, and is now selling the Cray XC30 series of supercomputers with NVIDIA Tesla K20X GPU accelerators and Intel Xeon Phi coprocessors. Us of the Intel Phi and NVIDIA Tesla GPU accelerators is a continuation of Cray’s support for innovative hybrid supercomputing technologies. In addition to now being offered in both the Cray XC30 and Cray XC30-AC systems, Intel Xeon Phi coprocessors and NVIDIA Tesla GPU accelerators are also available in the Cray CS300 line of cluster supercomputers.
“We designed the Cray XC30 supercomputer to be the realization of our Adaptive Supercomputing vision, which is about providing customers with a powerful, flexible tool for solving a multidisciplinary array of computing challenges,” said Peg Williams, Cray’s senior vice president of high performance computing systems. “Integrating diverse accelerator and coprocessor technologies into our XC30 systems gives our customers a variety of processing options for their demanding computational needs. Equally as important, Cray XC30 supercomputers feature an innovative software environment that allows our customers to optimize their use of diverse processing options for their unique applications.”