University of Connecticut Selects SGI Altix as Foundation of New Institute for Supercomputing and Visualization
November 15, 2005 09:04:18 (ET)
SEATTLE, Nov 15, 2005 /PRNewswire-FirstCall via COMTEX/ -- To initiate its long-planned Connecticut Institute for Supercomputing and Visualization, the University of Connecticut School of Engineering (Storrs Campus) purchased and recently installed two SGI(R) Altix(R) systems from Silicon Graphics (SGID, Trade).
An 8-processor SGI(R) Altix(R) 350 mid-range server with 8GB of memory will serve as a front-end to a 64-processor SGI(R) Altix(R) 3700 Bx2 supercomputer configured with 64GB of memory. Both systems will interact with the School's SGI(R) Onyx(R) visualization system and a 1TB direct-attached SGI(R) InfiniteStorage TP900 disk array to provide researchers with a seamless computational and visualization platform.
According to the School of Engineering's Ian Greenshields, Associate Dean-Academic Affairs, these two new systems will form the basis of a supercomputing facility that ultimately will become a national center of excellence in supercomputing research and applications.
The Altix systems were installed in early November at the Booth Engineering Center for Advanced Technology, a computing research center operated by the School of Engineering. Faculty from the School of Engineering and elsewhere in the University will rely upon the systems for a diversity of research applications. Following are several examples that illustrate how the Altix systems will be employed in computing-intensive research conducted by the School.
-- Biomolecular and Metabolic Engineering. Due to rapid advances in
sequencing technology, the genomes of numerous organisms have been made
readily available to the research community. Consequently, it is now
possible to create genome scale mathematical models of these organisms
and simulate their metabolic processes. These "in silico" models may
be used to identify metabolic pathways in a given organism that may be
engineered to carry out processes, such as development of a new
pharmaceutical drug. The models also may be used to identify potential
drug targets in pathogenic organisms, and to determine the best
potential strategy for treating a pathogen. The use of genome scale
modeling permits researchers to identify potential drug targets in
pathogenic organisms, as well as learn how to manipulate beneficial
organisms for the betterment of society. Implementation of these models
can be computationally intensive, especially when incorporating
stochastic phenomena or sub-cellular spatial heterogeneity. With the
Altix systems, the models can incorporate a substantial level of
detail, resulting in more accurate and realistic predictions.
-- Fuel Cells and Alternative Energy. Computational fluid dynamics (CFD)
is applied to understand highly coupled fluid dynamics, species
transport and electrochemical kinetics of fuel cells. These simulations
serve as a basis for understanding the limiting phenomena of fuel cell
operation, and for design improvements. Due to the intrinsic multi-
scale nature and complex multi-physics phenomena occurring in fuel
cells, these computations have very large memory and processing
requirements, and are thus very well suited for deployment on the Altix
systems.
-- Parachute Dynamics. One UConn researcher has been working with the U.S.
Army Soldier Systems Center (Natick, MA) on the development of
large-scale computational models for simulating the complex behavior of
parachute systems. The models typically exceed four million
computational elements, which mandates the use of massively parallel
computer systems. The ability to accurately simulate parachute system
behavior will reduce the amount of physical testing needed, and
therefore costs, associated with development of new parachute designs.
-- Uncertainty Analysis in Biological Systems. In seeking to understand
biological systems, researchers sometimes draw the analogy of a
chemical plant. In flexibility analysis, researchers seek to answer
questions such as: "During plant operation, can we ensure that process
constraints will be satisfied for a given range of uncertain parameter
values?" This analysis can help determine the adequacy of UConn
researchers' mathematical models or the accuracy of their process
constraints. If researchers view the metabolic processes occurring in
living cells -- for example, human cells -- as unit operations, then
conceptually, flexibility analysis could be applied to metabolic
processes. The scope of the problem is orders of magnitude larger than
this small example, and supercomputing will be a powerful tool in
solving flexibility analysis problems related to metabolic processes.
In addition to the use of the Altix systems in research, the School of Engineering has developed a wide-ranging set of outreach plans aimed at introducing 21st century computation to a wider audience.
The University's Altix run Novell(R) SUSE(R) Linux Enterprise Server, Version 9 and a range of applications, including Star-P(TM) from Interactive Supercomputing LLC. Star-P enables a client-server model in which a desktop client application is transparently linked with a powerful remote supercomputing server through a standard computer network. This greatly speeds up desktop computations, and allows larger problem sets to be handled.
Based on scalable Intel(R) Itanium(R) 2 Processors, SGI Altix systems are particularly well suited to UConn's scientific applications, due in large part to SGI's third-generation NUMAflex(TM) architecture. This unique global shared-memory architecture enables researchers to hold entire data sets in memory, allowing for faster and more interactive data analysis, and resulting in more incisive conclusions.
"The ambitious and visionary agenda for the new Connecticut Supercomputing and Visualization Institute involves research that requires some of the most advanced and scalable computing architectures available today," said Afshad Mistri, senior manager, advanced visualization and scientific markets segment, SGI. "SGI is delighted to provide the bedrock technology for this exciting new initiative, and we look forward to collaborating with the Institute as it pursues important breakthroughs in a broad range of disciplines."
About the University of Connecticut School of Engineering
The University of Connecticut is the state's flagship institution of higher learning, and a Land Grant and Sea Grant College as well as a Space Grant Consortium institution. Enrollment exceeds 27,000 students. Founded in 1901, the School of Engineering is the state's only public, accredited degree-granting engineering program and serves 2,100 undergraduate and graduate students. Additional details may be found at www.engr.uconn.edu.
SILICON GRAPHICS | The Source of Innovation and Discovery(TM)
SGI, also known as Silicon Graphics, Inc., is a leader in high-performance computing, visualization and storage. SGI's vision is to provide technology that enables the most significant scientific and creative breakthroughs of the 21st century. Whether it's sharing images to aid in brain surgery, finding oil more efficiently, studying global climate, providing technologies for homeland security and defense or enabling the transition from analog to digital broadcasting, SGI is dedicated to addressing the next class of challenges for scientific, engineering and creative users. With offices worldwide, the company is headquartered in Mountain View, Calif., and can be found on the Web at www.sgi.com.
NOTE: Silicon Graphics, SGI, Altix, Onyx, the SGI cube and the SGI logo are registered trademarks, and NUMAflex, and The Source of Innovation and Discovery is a trademark of Silicon Graphics, Inc., in the United States and/or other countries worldwide. Novell is a registered trademark, and SUSE is a trademark of Novell, Inc. in the United States and other countries. Intel and Itanium are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries. Linux is a registered trademark of Linus Torvalds in several countries. All other trademarks mentioned herein are the property of their respective owners.
MEDIA CONTACT
Marla Robinson
marlar@sgi.com
+1-256-773-2371
SGI PR Hotline
+1-650-933-7777
SGI PR Facsimile
+1-650-933-0283
SOURCE Silicon Graphics, Inc.
Marla Robinson, of SGI, +1-256-773-2371, or marlar@sgi.com, or SGI PR Hotline,
+1-650-933-7777, or SGI PR Facsimile, +1-650-933-0283
prnewswire.com |
