GPUs and the Future of Parallel Computing

by Stephen W. Keckler, William J. Dally, Brucek Khailany, Michael Garland, and David Glasco

Although Moore's law has continued to provide smaller semiconductor devices, the effective end of clock rate scaling and aggressive uniprocessor performance scaling has instigated mainstream computing to adopt parallel hardware and software. Today's landscape includes various parallel chip architectures with a range of core counts, capability per core, and energy per core. Driven by graphics applications' enormous appetite for both computation and bandwidth, GPUs have emerged as the dominant massively parallel architecture available to the masses. READ FULL ARTICLE (login required) »