Andrew Canning, a member of NERSC's Scientific Computing Group,
was part of an international team that won the 1998 Gordon Bell Prize for the best
achievement in high-performance computing. The winners of the prestigious award were
announced during SC98, an annual conference on high-performance computing and networking
held in Orlando, Florida. Canning's collaborators include scientists at Oak Ridge National
Laboratory (ORNL), the Pittsburgh Supercomputing Center and University of Bristol (UK).
The group was recognized for their modeling of metallic magnet atoms, which was run on
increasingly powerful Cray T3E supercomputers. They started with NERSC's 512-processor
machine and won the prize with a top performance of 657 Gigaflops (657 billion
calculations per second). The group later topped even that performance by achieving 1.02
Teraflops (trillions of calculations per second).
Funded as one of the U.S. Department of Energy's Grand Challenges, the group developed
the computer code to provide a better microscopic understanding of metallic magnetism,
which has applications in fields ranging from computer data storage to power generation
Also during the SC98 awards ceremony, NERSC's Phil Colella was
presented with the 1998 Sidney Fernbach award for his "outstanding contribution
in the application of high performance computers using innovative approaches."
NERSC's strong showing, says NERSC Division Director Horst Simon, clearly demonstrates
that Berkeley Lab is taking a lead role in the field of unclassified high-performance
"As the Department of Energy's national facility for computational science, we see
this achievement by the Grand Challenge team as a major breakthrough in high-performance
computing," Simon said. "Unlike other recently published records, this is a real
application running on an operational production machine and delivering real scientific
results. NERSC is proud to have been a partner in this effort."
Finally, the Gordon Bell Prize for best price/performance on a computer went to a
collaboration among universities and DOE national laboratories, including Brookhaven
National Laboratory and Fermi National Accelerator Laboratory. A member of this team, Greg
Kilcup, is a physics researcher at Ohio State University who has been a visiting
researcher at the Lab and is a long-time user of NERSC.
Under Secretary of Energy Ernest Moniz spoke at the conference and praised the
accomplishments of DOE scientists and supercomputing centers. "All of these
achievements are part of the Department of Energy's emphasis on taking supercomputing the
next big step forward," he said. "Revolutionary advances in computation and
simulation promise a new era for scientific discovery and technological innovation."
Although parallel supercomputers are the world's fastest computers, capable of
performing hundreds of billions of calculations per second, realizing their potential
often requires writing complex computer codes as well as reformulating the scientific
approach to problems, so that the codes scale up efficiently on these types of machines.
In developing the magnetism modeling code for parallel computers, the researchers were
forced to rethink their formulation of the basic physical phenomena.
"One of the goals of this project is to address critical materials problems on the
microstructural scale to better understand the properties of real materials," said
Malcolm Stocks, a scientist at Oak Ridge National Lab and leader of the project that won
the top honors. "A major focus of our research is to establish the relationship
between technical magnetic properties and microstructure based on fundamental physical
principles. The capability to design magnetic materials with specific and well-defined
properties is an essential component of the nation's technological future."
In May and June of this year the research team ran successively larger calculations on
a series of bigger and more powerful Cray supercomputers. After the simulation code
attained a speed of 276 Gflops on the Cray T3E-900 512-processor supercomputer at NERSC,
the group arranged for use of an even faster T3E-1200 at Cray Research Inc. and achieved
329 Gflops. They were then given dedicated time on a T3E-600 1024-processor machine at the
NASA Goddard Space Flight Center, which allowed them to perform crucial code development
work and testing before the final run at 657 Gflops on a T3E-1200 1024-processor machine
at a U.S. government site.
"These increases in the performance levels demonstrate both the power and the
capabilities of parallel computers; a code can be scaled up so that it not only runs
faster but allows us to study larger systems and new phenomena that cannot be studied on
smaller machines," said Andrew Canning.
The Gordon Bell Award work was part of a larger Department of Energy Grand Challenge
Project on Materials, Methods, Microstructure and Magnetism -- a collaboration between
ORNL, Ames Laboratory, Brookhaven National Laboratory, NERSC, and the Center for
Computational Science and the Computer Science and Mathematics Divisions at ORNL.
In addition to Canning and Stocks, the team included Balazs Ujfalussy, Xindong Wang,
Xiaoguang Zhang, Donald M. C. Nicholson, and William A. Shelton of Oak Ridge National
Laboratory; Yang Wang of the Pittsburgh Supercomputing Center; and B. L. Gyorffy of the H.
H. Wills Physics Laboratory, UK.