QLogic IB Powers Hyperion Supercomputer
From Lawrence Livermore National Laboratory
This is a Press Release edited by StorageNewsletter.com on November 26, 2008 at 3:37 pmQLogic Corp. announced that Lawrence Livermore National Laboratory has selected QLogic to provide the InfiniBand infrastructure for the Department of Energy’s (DOE) new large-scale supercomputer cluster project, known as ‘Hyperion.’
QLogic SilverStorm 9000 Series InfiniBand switches and 7200 Series InfiniBand adapters provide a reliable, high-speed communication path for hundreds of computational nodes on this supercomputing cluster project, which is already playing a critical role in the DOE’s National Nuclear Security Administration (NNSA) Stockpile Stewardship Program. The Hyperion multi-mission system currently tests new architectures to support the NNSA, and its role is being expanded to reach out to the commercial sector to validate new science-based applications.
The goal of the Hyperion project is to provide a development, testing and scaling environment for new cluster technologies and infrastructure critical to the mission requirements of NNSA’s Advanced Simulation and Computing program. This includes testing new hardware and software technologies and forming long-term relationships to ensure continuity in the development of new technologies for ever-larger systems over the long haul. Other industry leaders on the project include Cisco, DataDirect Networks, Dell, Intel, LSI Corporation, Red Hat, Sun and Supermicro.
“With QLogic and all the Hyperion team members, we are able to create a Linux-based cluster that none of us could have built on our own,” said Mark Seager, LLNL project leader. “This collaborative approach to doing business makes it possible to expand the market with small and large-scale clusters based on open standards. Working with QLogic we were able to make this complex simulation tool deployable with industry-standard hardware and software, and usable within typical industry development centers.”
“QLogic InfiniBand switches and adapters intelligently harness the power of the hundreds of computing nodes in this cluster so that it can deliver the massive amounts of productivity expected by power users like the Department of Energy,” said Frank Berry, vice president of marketing, QLogic Corp. “As the largest testbed of its kind in the world, Hyperion will provide its collaborators with the opportunity to develop and test applications at an unprecedented scale.”
The first half of Hyperion is now online and being used by the collaboration. When completed in March, 2009, the Hyperion cluster, located at Livermore Lab, will have at least 1,152 nodes with 9,216 cores, with about 100 teraFLOP/s peak, more than 9TB of memory, InfiniBand 4x DDR interconnectivity and access to over 47GB/s of RAID disk bandwidth. The Hyperion testbed includes two storage area networks (SANs), one based on ‘Data Center Ethernet’ and the other based on InfiniBand. Both SANs are currently deployed utilizing Lawrence Livermore’s unique ‘TorMesh’ topology.