FREMONT, CA and LEIPZIG, GERMANY--(Marketwired - Jun 17, 2013) - SGI (
SGI ICE X system Spirit is named after the B2 Stealth bomber and is enabling Gary Kedziora, an HPCMP PETTT Computational Materials Scientist, to implement algorithms that perform quantum mechanical simulations with computational time that scales linearly with respect to the number of atoms.
"Spirit is significantly faster than our previously available platform for running these linear-scaling calculations, which are becoming viable for production level work," said Kedziora. "This enables DoD computational material scientists to model larger and more complex materials using predictive quantum mechanical methods on thousands of SGI ICE X processor cores."
"Our customers are already flocking to the fastest system in the Department of Defense, finding that their applications are performing significantly better on the new system," stated Jeff Graham, the director of the AFRL DSRC. "SGI has delivered a system that is exceeding their standard benchmark performance on DoD applications by an average of over 27%, which translates into higher productivity for scientists and engineers across the DoD."
As one of the world's fastest distributed memory platforms, the SGI ICE X system enables Spirit with 144 TB of memory and one of the largest and fastest pure compute InfiniBand clusters. Running on standard Redhat Enterprise Linux®, the SGI ICE X Spirit Supercomputer is housed in 32 M-racks and includes 2,304 (IP-115 Gemini Twin) compute blades with cold-sink technology. It features 9,216 sockets (73,728 cores) powered by Intel ® Xeon ® E5 processor 2600 series operating at 2.6 GHz and boasts a theoretical peak performance of over 1.5 petaflops, further proving its ranking on the TOP500 list. The Spirit supercomputer also leverages the SGI InfiniteStorage™ 5500 providing 6.72 PB of storage and operating on the SGI Lustre® file system.
"By providing the technology solutions behind Spirit, SGI is further powering the work of the armed forces in the success of their missions and the safety of the men and women," said Jorge Titinger, president and CEO of SGI. "It is with great honor to see the results of the SGI ICE X system Spirit being included in the TOP500, further validating our work together and the effective supercomputing architecture we built to address complex HPC needs."
The SGI ICE X platform helped earn SGI the position in the Visionary quadrant in the latest Gartner Magic Quadrant for Blade Servers. With a tightly integrated Infiniband backbone, the ICE X system is ideal for tough workloads, and with advanced cooling and power solutions can supply petaflops of compute performance. In the M-cell configuration, the platform uses on-chip liquid cooling capability to allow warm-water cooling, eliminating the need for CRAC units, water chillers and towers. The liquid-cooled heat sink resides between the pair of node boards in each twin blade slot ("M-Rack" deployment) enabling the highest watt SKU support (e.g., 135W TDPs), utilizing a liquid-to-water heat exchanger that provisions the required quantity of flow for cooling and capturing 55 to 70% of heat generated on the node board. The M-Cell also helps control acoustics, allowing the team to have a conversation in their data center again. The SGI ICE X system provides the ultimate configuration flexibility including four different InfiniBand topologies to enable optimized performance, with the InfiniBand interconnect that now dominates the Top 100 of the TOP500 list. For example, the SGI ICE X supercomputer at Total was announced today as 11th on the Top 500 list. The SGI ICE X product is also available in a standard 19-inch "D-rack" configuration, using standard front-to-back airflow with more modest processor density.
ICE X products are available today and now comes with Intel® Xeon Phi™ coprocessors, which can be seen at Intel's Booth # 341+350 during this month's International Supercomputing Conference, June 16-20, 2013 in Leipzig, Germany. SGI will also be displaying the ICE X platform near the conference at NH Leipzig Messe, Fuggerstr. 2. D-04158 Leipzig, meeting rooms Schoenefeld Room and Ehrenberg Room.
About the High Performance Computing Modernization Program (HPCMP)
The HPCMP provides DoD laboratories and Test Centers HPC capabilities and expertise through partnerships with the Laboratories and Test Centers. HPCMP resources are used to provide military programs with a competitive advantage for the design, operation and sustainment of new weapons systems; for example, preparing U.S. aircraft for overseas deployments in Afghanistan and Iraq, and assisting in long-term weather predictions to plan humanitarian and military operations throughout the world. The Air Force Research Laboratory is such a partner with the HPCMP by providing a world class HPC Center committed to providing the resources necessary for DoD scientists and engineers to complete their research, development, testing and evaluation projects. Since the organization's inception in 1996 the AFRL DSRC has supported the warfighter by combining powerful computational resources, secure interconnects, and application software with renowned services, expertise and experience.
SGI, the trusted leader in technical computing, is focused on helping customers solve their most demanding business and technology challenges. Visit sgi.com for more information.
© 2013 Silicon Graphics International Corporation. SGI, the SGI logo, ICE, and InfiniteStorage are trademarks or registered trademarks of Silicon Graphics International Corp. or its subsidiaries in the United States and/or other countries. Intel and Xeon are registered trademarks of Intel Corporation. All other trademarks are property of their respective holders.
- Technology & Electronics
- Air Force Research Laboratory
- Department of Defense
- high performance computing
Ogilvy Public Relations