July 07, 2009
Company leads in performance efficiency; number one among open systems HPC deployments
FREMONT, Calif., July 7 -- SGI continues to showcase its leadership among providers of the world's most powerful, energy-efficient supercomputers. In the latest semi-annual ranking of the world's TOP500 Supercomputer Sites, the SGI Altix ICE cluster at the U.S. Army Research Lab was heralded as the industry's "greenest" supercomputer, as measured by performance efficiency. The results were presented by TOP500 researchers at the International Supercomputing Conference (ISC) on June 23 in Hamburg, Germany.
Performance efficiency, or "LINPACK efficiency," measures the ratio between maximum performance (RMax) and peak performance (RPeak). SGI dominated this new category with 12 of the 20 most performance-efficient supercomputers in the world.
SGI supercomputers continue to be well represented in the top open systems deployments, adding three new supercomputers to the prestigious TOP500 list.
Based on the published TOP500 data, SGI's Altix ICE cluster at the U.S. Army Research Lab also represents the first time that the performance efficiency of an x86 scalar-based system outperformed vector-based systems. The latter traditionally performs better due to its ability to quickly access single pieces of data; however, SGI Altix ICE's high-speed fabric, coupled with highly-tuned MPI libraries and Intel's new Nehalem processors, gave it the edge in this comparison.
"SGI has once again been recognized for continually pushing the boundaries of supercomputing and developing the most innovative, energy-efficient systems in the world," said Dr. Eng Lim Goh, SGI's chief technology officer. "We did not achieve these levels of performance efficiency by chance. We use the same hardware components available to competitors but, because of unique architectures like the Altix ICE, customers are able to obtain additional performance and energy efficiency from our systems."
SGI's "traditional" clusters and servers ensure the most demanding applications make dedicated use of all components: CPU, memory, bandwidth, OS, compilers and middleware layers, while guaranteeing a flexible and fast evolution path. This is due to the in-depth understanding of HPC applications and system optimization needed to exceed the demands of application workflows.
"Platforms based on Intel Xeon 5400 and 5500 series processors have been game-changing for supercomputing, and we've worked closely with SGI to develop powerful yet energy-efficient systems for the world's most demanding environments," said Richard Dracott, general manager of Intel's High Performance Computing Group.
The TOP500 Supercomputer Sites is compiled by Hans Meuer of the University of Mannheim, Germany; Erich Strohmaier and Horst Simon of NERSC/Lawrence Berkeley National Laboratory; and Jack Dongarra of the University of Tennessee, Knoxville.
SGI (NASDAQ:SGI) is a global leader in large-scale clustered computing, high-performance storage, HPC and datacenter enablement and services. SGI is dedicated to solving the IT industry's most demanding business and technology challenges. Visit www.sgi.com for more information.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?