Three grid computing experts at Binghamton University, State University of New York, say they now have the hardware and resources to take on questions that might never be answered without massively powerful computers. Associate Professor Michael Lewis and Assistant Professors Kenneth Chiu and Madhu Govindaraju, who form the “grid group” in Binghamton University’s Computer Science Department, are helping researchers in bioengineering, physics and other fields solve a huge array of computationally demanding problems. “We’re not going to cure cancer, but we’re going to help somebody cure cancer,” Chiu said. Grid computing relies on disperse and diverse resources to solve problems that require more computing power than is available from a single machine or from a local system. Grids also allow programmers to use different and more resources than they would otherwise have. While some fields might have a “holy grail,” an ultimate solution to the ultimate problem, grid computing has something more like a “holy scenario,” Chiu said. For instance, if a scientist is wondering whether there’s a connection between solar flares and autism, it would take a huge effort to find out. But since the data and the computational power exist, grid computing could make it possible to answer the question. “Ideally,” Chiu said, “grid computing allows scientists to test more hypotheses, which allows them to discover more stuff. This will increase the rate of scientific discovery.” Binghamton participates in four grids: the Open Science Grid, New York State Grid, PRAGMA Grid and TeraGrid. There’s also a 64-node high-performance cluster on campus. The grid group is working with researchers at Oakridge National Laboratory on a project related to “checkpointing,” which allows a computer to resume its computations where it left off after a shutdown or other problem. The 64-node cluster at Binghamton is serving as a testbed for that project, Govindaraju said. One of the most promising collaborations on the Binghamton campus is with Jacques Beaumont, associate professor of bioengineering, who’s studying heart arrhythmia. He’s especially interested in the cause of sudden cardiac death, in which a person who has had no previous symptoms suddenly dies of a cardiac arrest. The disease claims the lives of about 400,000 Americans each year. “How do we predict and diagnose this?” Beaumont asked. “How do we prevent it?” Beaumont hopes to answer those questions by building a model of the heart that would reveal who’s at risk for sudden cardiac death. Beaumont, working with researchers at Upstate Medical University and at the University of South Carolina, has created a computer program that simulates the generation of a cardiac beat, including the sequence of events leading to life threatening arrhythmia. The grid group has modified the program to take advantage of the huge processing power available through grid computing. Running the software on the grid may allow Beaumont and his colleagues to answer questions in days rather than years, he said. He dreams of a day when cardiologists will be able to forecast a heart attack the way a meteorologist can forecast a hurricane. “This,” Beaumont said, “is making the impossible possible.”