AEROSPACE

Linux Networx, The Linux Supercomputing Company, today announced that Computer Sciences Corporation (CSC) has ordered a Custom Supersystem for the NASA Center for Computational Sciences (NCCS) at the Goddard Space Flight Center. The new system is designed to dramatically increase throughput for applications ranging from studying weather and climate variability to simulating astrophysical phenomena. The system will supplement the NCCS architecture with improved price/performance and is designed to scale to as many as 40 trillion floating-point operations per second (TFLOPS) in its full configuration. "For over twenty years NASA Goddard has been at the forefront of adopting new supercomputing technologies and approaches," said Robert (Bo) Ewald, chief executive officer, Linux Networx. "We're honored to be selected as NASA Goddard's partner as they continue to advance their scientific computing capabilities." CSC and the NCCS selected a 128-node, 3.3 TFLOPS Supersystem with Intel dual-core Dempsey processors and an Infiniband Network from Silverstorm Technologies. This system includes a high-performance storage subsystem with 60 TBs of raw Data Direct Networks storage running the IBM General Parallel File System (GPFS). The storage subsystem will be tightly integrated with the supercomputer to further accelerate and optimize performance. This initial system will be the cornerstone of a cluster environment that is easily expandable to keep pace with the ever-increasing scientific requirements for computational capacity. In addition to powering the NCCS's application throughput and price/performance needs, the Custom Supersystem was chosen for its system reliability and the ability to incorporate new technologies over time. Linux Networx also offers unrivaled industry expertise and experience in meeting the needs of other government agencies. The NASA Goddard acquisition is the second major order by a government agency in three months. In February Linux Networx announced that the Department of Defense High Performance Computing Modernization Program placed the largest single order for Linux Supercomputers in the company's history. Linux Networx Custom Supersystems are especially well-suited for organizations that wish to pioneer new technologies, new programming models or new levels of supercomputing scale. As an experienced system partner, Linux Networx works with customers and technology partners to identify and integrate a variety of emerging and proven technologies into a production-ready system. Linux Networx leads the design, integration and testing of these next generation one-of-a-kind supercomputing clusters to the exacting specifications of our customers.