SCIENCE

Appro, the leading provider of supercomputing solutions, today announced the final deployment of Appro Supercomputing Clusters to Advanced Simulation and Computing (ASC) that integrates the work of the three National Nuclear Security Administration (NNSA) Defense Programs laboratories: Lawrence Livermore National Laboratory, Los Alamos National Laboratory and Sandia National Laboratories. This procurement was awarded to Appro last year for the TLCC07 program. Appro deployed to the three national laboratories 21 Scalable Units (SU) – 438 TFLOP/s, Linux high performance clusters and an additional 200 TFLOP/s of 10 SUs included as options. The Appro scalable clusters are being used principally to provide needed computational support to NNSA’s nuclear weapons programs, notably Stockpile Stewardship – the program to ensure the safety, security and reliability of the nation’s nuclear deterrent without nuclear testing. The additional 200 TFLOP/s capacity computing was needed for efforts as completion of the National Ignition Facility laser fusion project; and the program to extend the life of existing weapons in the stockpile. "The ability to provide a SU in multiple scalable clusters to all three national Lab sites with a 30-50% lower TCO was of great value to NNSA's effort to apply high performance computing to time-urgent national security challenges," said Mark Seager, ICCD ADH for Advanced Technology, Lawrence Livermore National Lab. "We needed a partner to bring strong technical skills to the table including an understanding of the HPC ecosystem as well as the project management skills necessary to work with a large number of component suppliers. Fielding 31 SUs (4,464 compute nodes with an aggregate performance of 620 TFLOP/s) at three Tri-Lab sites (LLNL, LANL, SNL) was a daunting project challenge that, in this case, was very well executed." “Lawrence Livermore Labs alone has up to 495 TFLOP/s of computing power spread over eighteen x86 Linux clusters for its scientific user research base. We are proud to say that 480 TFLOP/s of the eleven systems installed at LLNL are Appro supercomputing clusters representing the largest installed base of Linux Clusters in their site. We are also proud to be able to provide 638 TFLOP/s of computing power to the Three National Labs as a result of this contract. This is a very big accomplishment for Appro”; said John Lee, VP of Advanced Technology Solutions of Appro. “Quad-Core AMD Opteron processors offer outstanding scalability, especially in four socket systems. The benefits of Direct Connect Architecture – an integrated memory controller combined with HyperTransport technology delivered the node memory bandwidth performance that was critical to meet the three National Labs specification,” said David Rich, director of High Performance Computing, AMD.