dcsimg

University of Arizona Researchers Speed Time to Results With SGI Solutions

SUNNYVALE, Calif., July 16 /PRNewswire-FirstCall/ -- To replace current high-performance compute systems with systems of higher capacity and more capability, the Center for Computing and Information Technology (CCIT) at The University of Arizona (UA), working with the University's HPC Task Force, selected SGI (Nasdaq: SGIC) technology as a university-wide CCIT Research Computing resource. Installed in March, the new SGI(R) Altix(R) 4700 with 512-cores and 1TB of shared memory and the latest innovation in high-performance speed, SGI(R) RASC(TM) RC100 blade which leverages Field Programmable Gate Arrays (FPGAs), are being used to speed results across a wide spectrum of scientific and engineering research projects including life sciences, materials science, physical sciences, and mechanical engineering.

SUNNYVALE, Calif., July 16 /PRNewswire-FirstCall/ — To replace current high-performance compute systems with systems of higher capacity and more capability, the Center for Computing and Information Technology (CCIT) at The University of Arizona (UA), working with the University's HPC Task Force, selected SGI (Nasdaq: SGIC) technology as a university-wide CCIT Research Computing resource. Installed in March, the new SGI(R) Altix(R) 4700 with 512-cores and 1TB of shared memory and the latest innovation in high-performance speed, SGI(R) RASC(TM) RC100 blade which leverages Field Programmable Gate Arrays (FPGAs), are being used to speed results across a wide spectrum of scientific and engineering research projects including life sciences, materials science, physical sciences, and mechanical engineering.

In the University's recent tests, their new SGI Altix 4700 delivered up to six times the computing capacity of the University's retired HP system. In addition, because the Altix is configured with 1TB of shared memory, students and researchers can now run 20x higher resolution molecular dynamics simulations of entire viral protease instead of limiting their research to known binding sites. The Altix 4700 computing powerhouse is housed in just 5 racks, replacing the 17 racks of previous equipment, resulting in datacenter floor space, power, and cooling savings. The UA chose the Altix with a water cooling option which actually helps cool the datacenter itself. Today, over 120 research groups at the UA have requested use of the SGI Altix 4700 system and the number is growing.

One of the numerous colleges, departments and research labs leveraging the new SGI computing resource is the Biogeochemical Dynamics Laboratory (BGDL), in the Department of Geosciences. Joellen L. Russell, Assistant Professor of Geosciences, leads the lab and uses numerical models to simulate the Earth's climate system and carbon cycle. The lab uses global and regional climate models to predict the impact of changes in the physical climate system. One of the most exciting projects in the lab is simulations of the shift in the position of the Westerly Winds. The Westerlies have moved poleward and intensified over the last 30 years and are predicted in climate model simulations to move even further, decreasing wintertime precipitation in the Southwest and increasing the likelihood of drought and fire.

"SGI is the leader in providing optimized high performance computing solutions for weather and climate studies," said Russell, a widely-published author on ocean science and climate. "I am delighted that the UA has chosen SGI to assist us in building a world-class regional and global climate modeling infrastructure. My hope is that we'll be able to help provide information, strategies and solutions for Arizona's pressing problems in environmental sustainability."

The combination of SGI Altix systems with integrated FPGA technology accelerated BLAST searches in bioinformatics up to 20 times faster than their previous computing resource. The SGI RASC RC100 will be used to speed the UA's bioinformatics research, as well as a resource for the University's Arizona Research Laboratories (ARL) results in population genetics, evolutionary migration patterns, and bio-diversity.

"BLAST is heavily used by many of our Life Sciences research groups at the university and it is important to their research," said Dr. Michael Bruck, Assistant Director of Research Computing at the UA's Center for Computing and Information Technology (CCIT), which coordinates central research computer resources on campus. "The Mitrion-accelerated BLAST application gives them the ability to accelerate their searches and complete jobs sixteen-to-twenty times faster than on the systems they were running before. This increased throughput made possible through FPGA acceleration is definitely a productivity benefit to their research programs."

Research Professor Cari Soderlund, for example, heads the Arizona Genomics Computational Laboratory (AGCoL) that is currently working on the computational aspects of sequencing 30,000 genes for maize. In order to determine the sequence of genes, subsequences of about 800 base pairs (bp) are generated. To reconstruct the original gene sequence, the sub sequences are "Blasted" against each other to determine the overlapping sub-sequences. For the 700,000 sequences, according to Soderland, this would have taken a couple of weeks on a commodity PC with quad 2.8ghz processor with 8GB of RAM, and she looks forward to running an upcoming assembly on the SGI Altix system.

"As our lab is often processing large datasets, the speedup from the SGI Altix and RASC system would increase the turn-around time between the generation of biological data and computational analysis," said Soderlund. "It would also allow for more experimentation of parameters and algorithms for larger datasets."

"For life scientists, it is, by default, becoming a requirement to familiarize themselves with HPC systems," said Nirav Merchant, Director for Biotechnology Computing at ARL. "If you don't know how to manage high-throughput data, which requires high-throughput computing, you just cannot conduct your experiments. I regularly hold workshops for students on how to use HPC resources and they learn everything, from shared memory machines to grids, and how to prepare their data. The SGI Altix system, where traditional CPUs and RASC co-exist and work together with Mitrionics software at the core, is a solution that integrates well with our workflow."

"Our Mitrion-accelerated BLAST application running on the SGI RASC RC100 system delivers tremendous processing power with significantly lower power consumption — all in a turnkey solution," stated Anders Dellson, CEO of Mitrionics, Inc. "Together, SGI and Mitrionics are reaching new technology milestones as we deliver real-world, accelerated applications for FPGAs, to our customers using industry-standard hardware platforms and open source software."

In addition to projects mentioned, other science leveraging the power of the new SGI resources include the Department of Hydrology and Water Resources' research of the groundwater component to the hydrologic cycle and the Physics Department's numerical simulations of quantum chromodynamics (QCD). At the College of Engineering, researchers in the Aerodynamics and Mechanical Engineering Department are working on numerical investigation of transition in supersonic boundary layers; another College of Engineering project will use the SGI Altix system to simulate fluid structure interactions in aortic grafts, and yet another project will create computationally expansive patient-specific models of aneurysms and eyes for glaucoma research. At the College of Science, the Ochman lab works in the general field of bioinformatics and computational genomics of bacteria.

Through SGI's exclusive higher education partner, James River Technical, Inc., (JRTI), the UA installed two SGI Altix 4700 systems with a total of 1.136TB of memory. The main SGI Altix system is configured with 512 Intel(R) Itanium(R) 2 processors and 1TB memory, running Novell(R) SUSE(R) Enterprise Linux 10 with SGI(R) ProPack 5. The second SGI Altix 4700 runs 56 Intel(R) Itanium(R) 2 processors and has 112GB memory. The University also purchased two SGI(R) RASC RC100 blades with 4 FPGAs.

"The implementation of the Altix 4700 and the RASC technology at the UA is a shining example of the inherent benefits of matching the innovative research compute needs of a university with the right HPC system, " say Tom Mountcastle, President of JRTI. "Collectively, the UA research team had already realized a number of key milestones from the system and we fully expect these to continue in the future. This was a client that knew what they needed and wanted, and executed seamlessly on their plan. We are delighted to have them as a client."

"With the addition of the SGI Altix system and SGI RASC technology, the UA continues its dedication to academic excellence," said Michael Brown, Sciences Segment Manager, SGI. "While we expect to see significant speed-ups in the solution of existing scientific challenges, the really exciting transformation is that researchers will be able to use 512 CPUs and 1TB of memory to solve entire new classes of problems across a wide range of scientific disciplines."

SGI — Innovation for Results(TM)

SGI (Nasdaq: SGIC) is a leader in high-performance computing. SGI delivers a complete range of high-performance server and storage solutions along with industry-leading professional services and support that enable its customers to overcome the challenges of complex data-intensive workflows and accelerate breakthrough discoveries, innovation and information transformation. SGI solutions help customers solve their computing challenges whether it's enhancing the quality of life through drug research, designing and manufacturing safer and more efficient cars and airplanes, studying global climate, providing technologies for homeland security and defense, or helping enterprises manage large data. With offices worldwide, the company is headquartered in Sunnyvale, Calif., and can be found on the Web at http://www.sgi.com.

(C) 2007 SGI. All rights reserved. SGI, Atix, the SGI cube and the SGI logo are registered trademarks of SGI, and NUMAflex and RASC are trademarks in the United States and/or other countries worldwide. Xilinx and Virtex are registered trademarks of Xilinx, Inc. Linux is a registered trademark of Linus Torvalds in several countries. Intel, Itanium, and Xeon are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries. All other trademarks mentioned herein are the property of their respective owners.

This news release contains forward-looking statements regarding SGI technologies and third-party technologies that are subject to risks and uncertainties. These risks and uncertainties could cause actual results to differ materially from those described in such statements. The reader is cautioned not to rely unduly on these forward-looking statements, which are not a guarantee of future or current performance. Such risks and uncertainties include long-term program commitments, the performance of third parties, the sustained performance of current and future products, financing risks, the ability to integrate and support a complex technology solution involving multiple providers and users, and other risks detailed from time to time in the company's most recent SEC reports, including its reports on Form 10-K and Form 10-Q.

MEDIA CONTACT Marla Robinson marlar@sgi.com 256.773.2371 SGI PR hotline 650.933.7777 SGI PR FACSIMILE 650.933.0714

SOURCE SGI

Comments are closed.