Designing and optimizing compute kernels on NVIDIA GPUs
IBM Research, Austin, TX
In ASP-DAC ’09: Proceedings of the 2009 Asia and South Pacific Design Automation Conference (2009), pp. 224-229
@conference{jamsek2009designing,
title={Designing and optimizing compute kernels on NVIDIA GPUs},
author={Jamsek, D.A.},
booktitle={Proceedings of the 2009 Asia and South Pacific Design Automation Conference},
pages={224–229},
year={2009},
organization={IEEE Press}
}
The availability of high performance compute capability in NVIDIA GPUs has expanded their use in CAD environments. We will describe the basic compute models including host/device programming models, device multi-thread programming models, as well optimization and performance tuning techniques.
November 25, 2010 by hgpu