Aspects of GPU for general purpose high performance computing
Graduate School of Information Science and Technology, The University of Tokyo & JST, CREST
2009 Asia and South Pacific Design Automation Conference (2009) Volume: 5544, Publisher: Ieee, Pages: 216-223
@conference{suda2009aspects,
title={Aspects of GPU for general purpose high performance computing},
author={Suda, R. and Aoki, T. and Hirasawa, S. and Nukada, A. and Honda, H. and Matsuoka, S.},
booktitle={Proceedings of the 2009 Asia and South Pacific Design Automation Conference},
pages={216–223},
year={2009},
organization={IEEE Press}
}
We discuss hardware and software aspects of GPGPU, specifically focusing on NVIDIA cards and CUDA, from the viewpoints of parallel computing. The major weak points of GPU against newest supercomputers are identified to be and summarized as only four points: large SIMD vector length, small memory, absence of fast L2 cache, and high register spill penalty. As software concerns, we derive optimal scheduling algorithm for latency hiding of host-device data transfer, and discuss SPMD parallelism on GPUs.
March 23, 2011 by hgpu
