DON'T MISS
Aboard the Narrative Train III -- A Conference on Narrative Journalism, December 2 and 3

Vol. IV No. 15   ·   1 December 2000   

Search the Bridge

B.U. Bridge is published by the Boston University Office of University Relations.

Contact Us

Staff

We've got the power
Supercomputer inauguration gives researchers breakthrough technology

By Brian Fitzgerald

It’s not exactly the most impressive sight in the world, but appearances can be deceiving. The newest computer at the Center for Computational Science on Cummington Street looks like a pair of six-foot-tall speakers at a rock concert. The computer can’t amplify music, but that doesn’t mean it doesn’t rock: it can process as many as 96 billion calculations per second.

 
  Glenn Bresnahan, director of the Office of Information Technology's Computation and Visualization Group, next to the RS/6000 SP, the new supercomputer that doubles BU's supercomputing capacity. Photo by Albert L'Étiole
 

The RS/6000 SP, a version of IBM’s ASCI White – the most powerful supercomputer ever built – will be instrumental in research efforts in a variety of fields, including quantum physics, genome research, and the pursuit of a cure for Alzheimer’s disease.

"It doubles our supercomputing capacity," said the center’s director, CAS Physics Professor Claudio Rebbi, at a November 17 symposium celebrating the center’s 10th anniversary and the inauguration of the supercomputer, which BU acquired in August.

The symposium featured high-tech presentations by researchers from BU, the National Science Foundation, the National Center for Computing Applications, and IBM.

For example, Zhiping Weng, an ENG biomedical engineering assistant professor, demonstrated how high-performance computing is applied to her biology research by showing three-dimensional protein structures on a large screen. Solomon Eisenberg, an ENG associate professor of biomedical engineering, displayed computer models of cardiac defibrillation, which is the treatment of choice for patients with heart arrhythmias that don’t respond to drug therapy.

For a decade, the center has been a mecca for researchers, educators, and students who are pioneering advanced application of computing to problems in science and engineering. "Now, with the RS/6000 SP, the sky is the limit," said Glenn Bresnahan, director of the Office of Information Technology’s Computation and Visualization Group. The computer has performance-enhancing POWER3-II copper microprocessors, silicon switching technology, and the advanced software speed, scalability, and reliability needed to solve intricate challenges, from unlocking the secrets of disease to enabling breakthroughs in drug design and geological research.

It’s also not a bad chess player. In 1997, Deep Blue, an IBM RS/6000, defeated world champion Gary Kasparov.

The supercomputer will allow BU researchers to analyze the harmful buildup of particles – including senile plaques and neurofibrillary tangles – in brains of people with Alzheimer’s disease, a progressively degenerative disorder of the central nervous system. Pictures of these particles, taken by a confocal microscope, can be scrutinized using methods of statistical physics.

"The IBM SP lays the groundwork for a series of enhancements to our supercomputing capabilities over the next few years," said John Porter, Boston University’s vice president for systems and technology. Another dramatic increase in performance is expected in 2002, when the system will be upgraded with IBM’s POWER4 microprocessors.

"The drive to apply superior computation to a wide range of fields has led the center’s collaborators to explore ways in which to apply algorithms and hardware to address hugely diverse problems," said BU President Jon Westling. "IBM has entrusted Boston University to be the first academic institution to bring its RS/6000 SP computer online, which will keep us at the leading edge of advanced computing technology."

Westling encouraged the researchers at the symposium to use their creativity and intellect to tackle the most complex problems of our time. "In many ways, the work that you are doing and the technologies that you are using have become sort of the magician’s bag of tricks for the 21st century," he said. "Many believe that computational science will soon enable us to draw invaluable stock tips and cures for disease like so many silk scarves from a magician’s pocket. The visions bring to mind the irrepressible optimism that Jules Verne displayed a century ago as he peered from the edge of the 19th century and into another, and imagined that unclouded horizon of scientific progress. Turning such a promise into reality requires a lot of effort, steadfast support, and an imagination that is not altogether different from the imagination of a fine novelist or a poet."

       

7 December 2000
Boston University
Office of University Relations