HPCwire

Leading HPC
Solution Providers


























HPCwire >> Off the Wire

Minnesota Supercomputing Institute Gets New SGI Cluster


SUNNYVALE, Calif., May 23  -- On any given day, hundreds of researchers may vie for access to the computing systems housed at the University of Minnesota Supercomputing Institute for Digital Simulation and Advanced Computation. Even with some 2,000 processors at its disposal, the institute has sometimes struggled to keep up with growing demand.

Now, a massive new cluster solution from SGI will help ensure that current and emerging generations of processor-hungry applications won't slow down Minnesota scientists and engineers.

The 2,048-core SGI Altix XE 1300 cluster, expected to be installed this month, will transform the familiar submit-and-wait research experience into a vastly more interactive and productive one. By distributing complex applications across hundreds or thousands of the Altix XE cluster's processors, researchers will be able to complete calculations faster, run more iterations of a problem in less time, and refine their conclusions more quickly and effectively than ever before.

"This is an exciting and important acquisition for the Minnesota scientific community," said H. Birali Runesha, director, scientific computing and applications at the institute. "With the Altix XE cluster, we can push the capability of applications that once were running on a small number of cores. Now it will be easier to scale jobs and achieve faster turn-around times. In fact, it might become routine to see applications scaled across 1,000 cores or more."

For some researchers, the new cluster will revolutionize their productivity. "This new system, with its dramatically faster job turn-arounds, will enable studies that we just can't pursue today," added Runesha.

Outfitted with more than 4 terabytes of memory across 256 compute nodes, Minnesota's new Altix XE cluster will drive research in physical, biological, medical, mathematical and computing sciences, in addition to engineering studies and academic-industry collaboration.

For example, increasingly complex computational fluid dynamics (CFD) and astrophysics applications will now be able to leverage so many processors that engineers can interact with simulations on the university's 8-foot-by-6-foot PowerWall from the Laboratory for Computational Science and Engineering. With the Altix XE cluster bearing the computational workload and feeding results to the PowerWall's visualization system, researchers can, for instance, adjust various parameters and assumptions as they employ a grid of more than 6 billion cells to simulate entrainment of unburned hydrogen fuel into the convection zone above the helium burning shell in a dying giant star.

This interactivity brings a new dimension to research, enabling scientists to experience breakthroughs at the speed of thought instead of waiting through multiple iterations of generating new scenarios, running batch simulations and visualizing their results.

"A cluster this extensive will help Minnesota researchers pursue the kind of cutting-edge science required to write strong grant proposals for the National Science Foundation and other important sources of funding," added Runesha. "No longer will these researchers be limited by their resources."

Minnesota Goes Quad-Core with Altix XE

Working with SGI and James River Technical, Inc., SGI's designated partner for higher education and research, institute administrators selected the Altix XE solution, based on Quad-Core Intel Xeon processors, after evaluating cluster solutions from multiple providers. From the outset, Minnesota targeted Linux cluster solutions to augment their installed base of large-scale supercomputers and servers. "No matter how large your systems, with enough users running enough applications," explained Runesha, "it's still possible to run out of processors."

To pinpoint the best-performing solution, Minnesota ran a customized benchmark suite that measured capabilities vital to a range of scientific and engineering applications. The suite tested functions used in well-known software -- such as GAMESS (General Atomic and Molecular Electronic Structure System) for computational chemistry, and BLAST (Basic Local Alignment Search Tool) for life sciences -- along with several in-house codes. The benchmark results showed that the Quad-Core Altix XE 1300 system delivered superior price/performance overall compared to Dual-Core systems, while minimizing the need for additional infrastructure.

SGI's ability to provide a full Double Data Rate (DDR) interconnect in a non-blocking topology also provides an edge to researchers. Non-blocking solutions maximize the performance of systems used by multiple users and applications. That's an important advantage for the institute, which currently serves some 2,000 users.

"Large, shared-memory Altix systems are fixtures at leading research institutions, including the University of Minnesota," said Bill Mannel, director of systems marketing, SGI. "But with multi-core SGI Altix XE clusters, even the most cost-conscious institutions can afford unique SGI technology, expertise and support, while pursuing a classic 'scale-out' cluster strategy."

The water-cooled Altix XE cluster also lowers the cost of owning and operating the cluster in an environment already dense with computing systems. "Since our facility, like many others across the country, is approaching either power or cooling limits, the SGI water-cooled option was very attractive to us," said Brian Ropers-Huilman, director, systems and operations at the institute.

Minnesota's system also is linked via a high-bandwidth InfiniBand connection and runs SUSE Linux Enterprise Server 10 from Novell. The SGI Altix XE server and cluster platform offers superior performance and energy efficiency at a breakthrough price point. Unlike other cluster offerings, SGI Altix XE combines the advanced Intel Xeon processor architecture delivered in a fully factory integrated cluster solution and backed by SGI's industry-leading service and support. The Altix XE platform delivers leading density without sacrificing performance or functionality.

"We are very excited about the selection and installation of an Altix XE system of this magnitude at the University of Minnesota," said Tom Mountcastle, president, James River Technical. "Over the years, Minnesota has continued to be at the forefront of technology implementations designed to expedite research and time to discovery. They have a clear understanding of the tools that are needed in their scientific community and the Altix XE will serve them well in this pursuit."

About SGI

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.

-----

Source: SGI


Article Tools

  • Print This Article

Share & Save Options

Discussion

There are 0 discussion items posted.  

Sponsored Links



Feature Articles

Nexsan Looks to Scare Up HPC Customers With Storage Beast

Even though the cost of servers still dominates the datacenter budget, storage is actually on a steeper growth curve. HPC storage, in particular, is being singled out as high-growth opportunity. Vendors are scrambling to keep up.
Read More...

The Week in Review

Google datacenters most energy efficient; Cluster Resources to demo Moab Hybrid Cluster; Red Hat Linux releases HPC distro. John West recaps those stories and more in our weekly wrap-up.
Read More...

Saudi Arabia Buys Some Big Iron

Last week, IBM and King Abdullah University of Science and Technology announced a collaboration to build "Shaheen," a 222 teraflop Blue Gene/P supercomputer. When deployed in 2009, it will represent the most powerful computer in the Middle East and one of the top systems in the world.
Read More...

Top Headlines

Oracle and HP's Database Machine Predicated on Voltaire

Oct 06 | The Register | Does the HP Oracle Database Machine represent InfiniBand's big chance to break out its HPC niche? Read more...

3D Imaging Spreads to Fashion and Beyond

Oct 06 | BusinessWeek | A body scan can save a lot of time in the fitting room, and fields from medicine to architecture are adopting 3D computing applications. Read more...

Structural Engineers and Computer Scientists Hope to Integrate Disciplines to 'Revolutionize Building Construction'

Oct 03 | UCSD News | Despite the evolution of computer science over the past 30 years, structural engineering -- hindered by a reluctance to adapt to digital innovations -- has remained relatively unchanged as a discipline. Read more...

Credit Crisis Spreads a Pall Over Silicon Valley

Oct 02 | New York Times | Silcon Valley is starting to feel the effects of the credit crunch. Read more...

Google: 'The World's Most Efficient Data Centers'

Oct 01 | Data Center Knowledge | Google today disclosed details of its data center energy usage, confirming that it operates some of the most efficient facilities in the world. Read more...

Featured Whitepapers

Panasas® Tiered Parity™ Architecture

Sep 04 | | Disk drives are approximately 250 times denser today than a decade ago. This is good news for users who are creating, manipulating and storing more data than ever before. It gives them an opportunity to derive more value from their stored data and lowers the capital acquisition and operating expense associated with that data.

Multimedia

Video White Paper: Architecting a Better Network Storage Solution

BlueArc's Titan architecture represents an evolutionary step in file servers by creating a hardware-based file system that can scale bandwidth, IOPS, and overall data capacity well beyond conventional software-based devices. With its ability to virtualize a massive storage pool of up to four usable petabytes of tiered storage, Titan can scale with growing data requirements, offering a competitive advantage for businesses, researchers, or other enterprises seeking to better manage data growth while still ensuring optimal performance.

High Performance on Wall Street

Newsletters

Stay informed! Subscribe to HPCWire email Newsletters.

Get updates and insights on the High Productivity Computing industry delivered driectly to your inbox.






Featured Events

Harvard Summit 2008
LCI Workshop
SIFMA
HP-CAST
2008 Virtualization Conference & Expo
Symposium 2009

HPC Job Bank