Hanspeter Pfister
GPUs for High-Throughput Science

Wednesday, December 15, 2010
Building 3 Auditorium - 11:00 AM
(Coffee at 10:30 AM)

How is the brain wired? How did the universe start? How can we predict and prevent heart attacks? These are some of the great scientific challenges of our times, and answering them requires bigger scientific instruments, increasingly precise imaging equipment, and ever more complex computer simulations. The traditional model is to process the data on a remote supercomputer. However, low data transmission rates, high energy consumption, and the high price of large parallel machines are obstacles for many scientists. In this talk I will suggest that commodity high-throughput GPU computing is enabling high-throughput science, where we process massive data streams efficiently and analyze them rapidly, all the way from the instrument to the desktop. I will present an overview of three projects at Harvard that leverage GPUs for high-throughput science, ranging from neuroscience and radio astronomy to computational fluid-flow simulations.

Hanspeter Pfister is Gordon McKay Professor of the Practice in the School of Engineering and Applied Sciences at Harvard University. His research lies at the intersection of visualization, computer graphics, and computer vision. Before joining Harvard he worked for 11 years at Mitsubishi Electric Research Laboratories where he was most recently Associate Director and Senior Research Scientist. Pfister has a Ph.D. in Computer Science from the State University of New York at Stony Brook and an M.S. in Electrical Engineering from the Swiss Federal Institute of Technology, Zurich, Switzerland. The web page for his group is http://gvi.seas.harvard.edu/pfister

IS&T Colloquium Committee Host: Tony Gualtieri

Sign language interpreter upon request: 301-286-7040
Request future announcements