Computer Engineering and Applications ›› 2012, Vol. 48 ›› Issue (22): 84-87.

Previous Articles     Next Articles

GPU programming and acceleration of Laplace growth model

ZHANG Xuezhi1, QI Ji2, LIN Ping2   

  1. 1.School of Physics Science and Technology, Lanzhou University, Lanzhou 730000, China
    2.Institute of Modern Physics, Chinese Academy of Science, Lanzhou 730000, China
  • Online:2012-08-01 Published:2012-08-06

拉普拉斯生长的GPU实现与硬件加速

张学智1,齐  记2,林  平2   

  1. 1.兰州大学 物理科学与技术学院,兰州 730000
    2.中国科学院 近代物理研究所,兰州 730000

Abstract: A 2D hcp grid Laplace random walk model program based on CUDA (Compute Unified Device Architecture) is coded and tested. The computing efficiency result compared to the corresponding CPU (Central Processing Unit) is given and comparing diagrams are presented.

Key words: Laplace growth, random walk, Compute Unified Device Architecture(CUDA)

摘要: 编写实现了六方点阵中拉普拉斯模型的随机行走雪花生长的CUDA(Compute Unified Device Architecture)程序,分析了GPU(Graphic Processing Unit)随机行走计算效率,对比了拉普拉斯模型随机行走雪花生长的GPU与CPU加速特性,给出了该模型随机计算环境尺寸增加的用时增长曲线。

关键词: 拉普拉斯模型, 随机行走, 通用并行计算架构(CUDA)