Supercomputing researchers have recently broken the petaflop barrier, or 10^15 floating operations per second. The next big barrier is to increase that power by 1000 times to exascale. Rather than focus on one big monolithic math machine, the goal is to incorporate the technological breakthroughs required to reach exascale computing and make them available in smaller machines.
With current hardware, this goal is unlikely to be reached even by 2020. Currently, most power and complexity is spent on data transfer from memory and across chip boundaries. Power and heat limitations mean that supercomputers can no longer count on increasing clock speeds. In order to hit the aggressive goal of 2015 for exascale computing, components must be designed for more efficient memory transport and be less concerned simply with keeping the math units busy.
Peter Kogge was with IBM, Federal Systems Division, from 1968 until 1994, and was appointed an IEEE Fellow in 1990, and an IBM Fellow in 1993. In 1977 he was a Visiting Professor in the ECE Dept. at the University of Massachusetts, Amherst. From 1977 through 1994, he was also an Adjunct Professor in the Computer Science Dept. of the State University of New York at Binghamton. In August, 1994 he joined the University of Notre Dame as first holder of the endowed McCourtney Chair in Computer Science and Engineering (CSE). Starting in the summer of 1997, he has been a Distinguished Visiting Scientist at the Center for Integrated Space Microsystems at JPL. He is also the Research Thrust Leader for Architecture in Notre Dame's Center for Nano Science and Technology. For the 2000-2001 academic year he was the Interim Schubmehl-Prein Chairman of the CSE Dept. at Notre Dame. Starting in August, 2001 he is the Associate Dean for Research, College of Engineering at Notre Dame. Starting in the fall of 2003, he also is a Concurrent Professor of Electrical Engineering at Notre Dame.
This free podcast is from our IEEE Spectrum Radio series.