Parallel computing experiences with CUDA

被引：330

作者：

Garland, Michael ^{[1
]}

Le Grand, Scott ^{[1
]}

Nickolls, John ^{[1
]}

Anderson, Joshua ^{[2
]}

Hardwick, Jim ^{[3
]}

Morton, Scott

Phillips, Everett

Zhang, Yao ^{[4
]}

Volkov, Vasily ^{[5
]}

机构：

[1] NVIDIA, Santa Clara, CA 95050 USA

[2] Iowa State Univ, Ames, IA 50011 USA

[3] TechniScan Med Syst, Salt Lake City, UT USA

[4] Univ Calif Davis, Dept Elect & Comp Engn, Davis, CA 95616 USA

[5] Univ Calif Berkeley, Dept Elect Engn & Comp Sci, Berkeley, CA 94720 USA

来源：

IEEE MICRO | 2008年 / 28卷 / 04期

关键词：

D O I：

10.1109/MM.2008.57

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

The CUDA programming model provides a straightforward means of describing inherently parallel computations, and NVIDIA's Tesla GPU architecture delivers high computational throughput on massively parallel problems. This article surveys experiences gained in applying CUDA to a diverse set of problems and the parallel speedups over sequential codes running on traditional CPU architectures attained by executing key computations on the GPU.

引用

页码：13 / 27

页数：15

共 22 条

[1]

ANDERSON JA, 2008, J CHEM PHYS, V128

[2] General purpose molecular dynamics simulations fully implemented on graphics processing units [J].