Computers based on the graphical processing unit (GPU) chips that power today's computer games are increasingly being used in place of supercomputers for many technical research tasks.
Researchers at Temple University's Institute for Computational Molecular Science, for example, used a server with a cluster of six specialized GPUs to continue their work modeling physical molecular interactions without using one of the school's supercomputers. "To do this number of calculations for the phase diagrams, you would not have the computer power locally, so you would have to apply to the supercomputing centers, but our research is not necessarily attractive to them, so we would not get the time," says Temple professor Axel Kohlmeyer.
For many difficult problems, GPU clusters can deliver up to 100 times more calculations than a typical system with the same number of processors, take up less space, and use much less power. "Two factors are of interest: Whether there is a lot of data involved and whether there is a lot of computation involved," says Nvidia's Sumit Gupta. "If those are true, the GPUs can likely be involved."
View Full Article
Abstracts Copyright © 2010 Information Inc., Bethesda, Maryland, USA