GOVERNMENT

MOUNTAIN VIEW, Calif. -- SGI announced that the Center for Computational Sciences at the Department of Energy’s Oak Ridge National Laboratory has purchased a new, 256-processor SGI Altix™ 3000 system. The new system will enable a powerful new class of applications for ORNL researchers, whose far-reaching efforts strengthen the nation’s leadership in the sciences, clean energy management and production, environmental protection, and homeland security. “Acquiring technology of this class for use in advanced research is a major collaborative achievement between scientists at our universities and national laboratories and advanced scientific computation vendors,” said Dr. Raymond L. Orbach, director of the Department of Energy’s Office of Science. “We look forward to collaborating further with SGI to ensure that we continue to benefit from the most sophisticated computing systems using their global shared memory capabilities. The search for knowledge that will impact the human experience is a magnificent enterprise, one the Office of Science is proud to foster.” “The important work pursued every day at the ORNL Center for Computational Sciences involves data sets of extraordinary size and complexity that only our newly introduced Altix 3000 system can handle,” said Bob Bishop, chairman and CEO of SGI. “We are delighted to provide this great institution with a system worthy of its acclaimed efforts. And by working closely with customers like ORNL, SGI continues to gain invaluable insights into the advanced science applications that will drive the future economy of the nation.” With 256 new Intel® Itanium® 2 processors, 2TB of global shared memory, and 1.5 TFLOPS of computational power at their disposal, ORNL researchers can simulate and analyze data sets of extraordinary size and complexity. The groundbreaking capabilities of the SGI Altix 3000 system will help ORNL drive new generations of scientific applications hungry for increased computing power and capacity. The applications include those used in computational biology and genetic research, as well as climate modeling, in which researchers project the potential long-term impact of such environmental threats as pollution and ozone depletion. The Center for Computational Sciences was established at ORNL in 1991 to evaluate new computer architectures for high-end computing. CCS works closely with industry, academia and sister labs in developing new scalable scientific applications to exploit these new high-end computing architectures to enable advances in a broad range of science and engineering. “We are very pleased that there are multiple vendors focused on architectures for high-end computing. The SGI Altix 3000 system, with its large globally addressable shared memory, offers a capability that complements our current and planned activities in high-end computing for science and engineering,” said Thomas Zacharia, associate lab director for ORNL. “Together with our core university partners – Georgia Tech, Virginia Tech, the University of Virginia, the University of Tennessee, Florida State, Duke University and North Carolina State University – and our collaborators in DOE labs, academia and industry, we plan to evaluate and help develop the potential of this new architecture for complex scientific and engineering simulations in biology, climate, materials, and fusion, among others.” To help manage the extreme data flows expected from using large data sets on the SGI Altix 3000 system, ORNL also has purchased more than 12TB of SGI® Total Performance 9100 (TP9100) RAID storage. The SGI® TP9100 is a high-performance 2Gbit Fibre Channel RAID array that has been certified on the SGI Altix 3000 platform for use as direct-attached storage or as fabric-attached storage in a storage area network. The SGI Altix 3000 system’s open standards-based Linux operating environment also dovetails with the laboratory’s growing interest in moving high-performance computing applications to proven large data Linux platforms. The SGI Altix 3000 family of servers and superclusters combines SGI® supercomputing architecture with Intel Itanium 2 processors and the Linux operating system. SGI Altix 3000 is recognized as the first Linux cluster that scales up to 64 processors within each node and the first cluster ever to allow global shared memory access across nodes. Inspired by the success of the SGI Altix family and the powerful combination of standard Linux on Intel Itanium family-based processors, more than 50 high-performance manufacturing, science, energy and environmental applications have been ported by their commercial developers to the 64-bit Linux environment, more than half of which have certified and optimized for the platform.