NASA Supercomputer Ranks Among World's Fastest

NASA's newest supercomputer at Ames Research Center, Moffett Field, Calif., has garnered the number three spot on the Top500 list of the world's most powerful computers. The announcement was made Nov. 17, 2008 at the International Conference for High-Performance Computing, Networking, Storage, and Analysis (SC08) in Austin, Texas.

The Pleiades supercomputer is an SGI Altix ICE system with 12,800 Intel Xeon quad-core processors (51,200 cores, 100 racks) running at 487 trillion floating point operations per second (teraflops) on the LINPACK benchmark, the industry standard for measuring a system's floating point computing power.

One of the most powerful general-purpose supercomputers ever built, Pleiades also features the world's largest InfiniBand interconnect network.

The LINPACK run also measured electrical power consumption-an increasingly important consideration in high-end computing. Using a total of 2.09 megawatts, or 233 megaflops per watt, Pleiades is among the most energy-efficient supercomputers in the world.

"Pleiades represents a significant engineering achievement in several ways," said William Thigpen, Pleiades project manager at the NASA Advanced Supercomputing (NAS) Division at Ames.


In addition to its power and InfiniBand record, "Pleiades can run NASA codes with minimal modifications, and is compatible with standard desktop engineering workstations so our users can migrate codes easily from their desktops. Users from all key mission areas will have an enormous resource to meet their critical milestones," Thigpen added.

Among the scientific and engineering projects accepted for computer time on Pleiades:

- Extensive simulations of large computational problems for future space vehicle design;

- Development of increasingly detailed models of large-scale dark matter halos and galaxy evolution;

- Running coupled atmosphere-ocean models to assess decadal climate prediction skill for the Intergovernmental Panel on Climate Change.

"Over a record-making few weeks, NASA has again deployed one of the most powerful supercomputers in the world," said Eng Lim Goh, chief technology officer at SGI.

"In the race to achieve superior computational power, NASA's knowledge of rapid yet productive deployment is a rare advantage. We are proud to be part of NASA's ongoing journey to show the world what is possible."

"We look forward to seeing the science breakthroughs enabled by Pleiades," said Stephen Wheat, senior director for high performance computing at Intel Corp. "It's rewarding to see that the performance features of the Intel Xeon quad-core processors meet the growing computational challenges of the nation's space program."

The InfiniBand fabric interconnecting Pleiades' 6,400 nodes requires more than 20 miles of double data rate cabling. InfiniBand is also used as the primary local-area network backbone that interconnects computing, storage, and visualization systems, and to facilitate cross-system data file access.

This enables scientific visualization and data analysis to execute concurrently as computer jobs run, producing ultra-high-fidelity results for the enormous datasets used in NASA mission projects.

Pleiades was acquired to augment the space agency's Columbia supercomputer (ranked No. 2 on the Top500 list in November 2004) in supporting NASA's four key mission areas: aeronautics research, exploration systems, science, and space operations.

source : http://www.spacedaily.com/

Continue Reading...

Microsoft, Intel research parallel computing with US universities

Microsoft and Intel said Tuesday they are teaming with US universities to unleash the mighty potential of multi-core computer chips.

Microsoft and Intel will jointly spend 20 million dollars over five years to fund Universal Parallel Computing Research Centers at the University of California, Berkeley, and the University of Illinois at Urbana-Champaign.

A recent trend is to increase computing power but reduce electricity use and heat production by crafting multiple processors, or computer brains, into each chip.

Designing software and support architecture that best enables such chips to divide tasks equally among the brains, or cores, in ways that let them simultaneously tend to jobs and maximize computing speeds has proved daunting.

"It is important for industry to work in tandem with academia to unleash the immense power of parallel computing," said Microsoft Research vice president Tony Hey.

"Working jointly with industry and academia, we plan to explore the next generation of hardware and software to unlock the promise and the power of parallel computing and enable a change in the way people use technology."

Intel and rival Advanced Micro Devices already market chips with two or four cores and say it is likely the number will rise. Intel researchers have made an 80-core research processor.

"We're quickly moving the computing industry to a many-core world," said Intel Research director Andrew Chien.

"Working with Microsoft and these two prestigious universities will help catalyze the long-term breakthroughs that are needed to enable dramatic new applications for the mainstream user."

Harnessing the power of multi-core chips will let computers "bridge the physical world with the virtual," according to Chien.

He expects "efficient and robust" applications for digital media, data analysis, and Internet-enabled mobile devices.

Predicted research breakthroughs include software enabling people's mobile telephone to recognize faces of approaching acquaintances and whisper their names to users.

Another foreseeable application is described as voice recognition software so accurate it could be used to record witness testimony in courtroom proceedings.

"This is a once-in-a-career opportunity to recast the foundations of information technology and influence the entire IT industry for decades to come," said UC Berkeley professor of computer sciences David Patterson.

The universities will have to provide millions of dollars in funding for the centers in what is said to be an unprecedented parallel computing research alliance.

source : http://www.spacedaily.com/

Continue Reading...

Jaguar Upgrade Brings ORNL Closer To Petascale Computing

Upgrades to Oak Ridge National Laboratory's Jaguar supercomputer have more than doubled its performance, increasing the system's ability to deliver far-reaching advances in climate studies, energy research, and a wide range of sciences.

The system recently completed acceptance testing, running applications in climate science, quantum chemistry, combustion science, materials science, nanoscience, fusion science, and astrophysics, as well as benchmarking applications that test supercomputing performance.

The Jaguar system, a Cray XT4 located at ORNL's National Center for Computational Sciences, now uses more than 31,000 processing cores to deliver up to 263 trillion calculations a second (or 263 teraflops).

"The Department of Energy's Leadership Computing Facility is putting unprecedented computing power in the hands of leading scientists to enable the next breakthroughs in science and technology," said ORNL Director Thom Mason. "This upgrade is an essential step along that path, bringing us ever closer to the era of petascale computing [systems capable of thousands of trillions of calculations per second]."

Jaguar was among the most powerful computing systems within DOE's Office of Science even before the recent upgrade and has delivered extraordinary results across a broad range of computational sciences.

"The leadership capability at Oak Ridge has been delivering real scientific results," said Michael Strayer, associate director for advanced scientific computing research in the DOE Office of Science.

"Benoit Roux of the University of Chicago used Jaguar to simulate in unprecedented detail the voltage-gated potassium channel, a membrane protein that responds to spikes of electricity by changing shape to allow potassium ions to enter a cell. This work has the potential to help us understand and control certain forms of cardiovascular and neurological disease."

Climate scientists are calculating the potential consequences of greenhouse gas emissions and the potential benefits of limiting these emissions. Combustion scientists are modeling the most efficient designs for engines that use fossil fuels and biofuels.

Fusion researchers are using the system to lead the way toward a clean and plentiful source of electricity. Physicists are exploring the secrets of the universe, illuminating its most elusive mysteries. And materials scientists are searching for the next revolution in technology.

"This is an important advancement," said Thomas Zacharia, ORNL associate laboratory director for computing and computational sciences.

"Leading researchers need many orders of magnitude more computing power and infrastructure than we can yet provide, and they have shown us how they will use these new resources, whether it be to predict the consequences of climate change at the regional level, design new materials with predetermined properties, discover new chemical catalysts, explore more efficient ways to manufacture biofuels, or simulate all important aspects of new reactor designs."

"The U.S. Department of Energy and its Oak Ridge National Laboratory have been making huge strides in providing more and more simulation capabilities to advance some of the world's most important scientific and engineering research-and invaluable partners with Cray to push the leading edge of supercomputing," said Peter Ungaro, president and CEO of Cray.

"This upgrade is another big milestone in leadership computing and we, along with many others around the world, are looking forward to learning about the scientific breakthroughs that are borne as a result of this powerful new computing capability."

With its new power, Jaguar will be able to double its contribution to DOE's Innovative and Novel Computational Impact on Theory and Experiment program, which is revolutionizing key areas of science by facilitating the world's most challenging computer simulations.

The NCCS will host 30 INCITE projects in 2008 from universities, private industry, and government research laboratories, contributing more than 140 million processor hours on Jaguar.

source : http://www.spacedaily.com/

Continue Reading...