耗散粒子动力学GPU并行计算研究

林晨森; 陈硕; 李启良; 杨志刚

doi:10.7498/aps.63.104702

摘要

研究了耗散粒子动力学基于计算统一设备架构的图形处理器（GPU）并行计算的实施. 对其中涉及的算法映射模型、Cell-List法数组的并行化更新、随机数生成、存储器访问优化、负载平衡等进行了详细的讨论. 进一步模拟了Poiseuille流动和突扩突缩流动，从而验证了GPU计算结果的正确性. 计算结果表明，相对于基于中央处理器的串行计算，在耗散粒子动力学中实施GPU 并行计算可以获得约20倍的加速比.

关键词:

Abstract

In this paper, the graphic processing unit (GPU) parallel computing of dissipative particle dynamics (DPD) based on compute unified device architecture is carried out. Some issues involved, such as thread mapping, parallel cell-list array updating, generating pseudo-random number on GPU, memory access optimization and loading balancing are discussed in detail. Furthermore, Poiseuille flow and suddenly contracting and expanding flow are simulated to verify the correctness of GPU parallel computing. The results of GPU parallel computing of DPD show that the speedup ratio is about 20 times compared with central processing unit serial computing.

Keywords:

作者及机构信息

1.
同济大学航空航天与力学学院, 上海 200092;

2.
同济大学上海地面交通工具风洞中心, 上海 201804

基金项目: 中央高等学校基本科研基金（批准号：125065）、国家自然科学基金（批准号：51276130，10872152）和教育部高等学校博士学科点专项科研基金（批准号：20120072110037）资助的课题.

Authors and contacts

1.
School of Aerospace Engineering and Applied Mechanics, Tongji University, Shanghai 200092, China;

2.
Shanghai Automotive Wind Tunnel Center, Tongji University, Shanghai 201804, China

Funds: Project supported by the Fundamental Scientific Research Foundation for the Central Universities of China (Grant No. 125065), the National Natural Science Foundation of China (Grant Nos. 51276130, 10872152), and the Specialized Research Fund for the Doctoral Program of Higher Education of China (Grant No. 20120072110037).

参考文献

[1]	Zhang M K, Chen S, Shang Z 2012 Acta Phys. Sin. 61 034701 (in Chinese) [张明焜, 陈硕, 尚智 2012 物理学报 61 034701]
[2]	Liu H T, Liu M B, Chang J Z, Su T X 2013 Acta Phys. Sin. 62 064705 (in Chinese) [刘汉涛, 刘谋斌, 常建忠, 苏铁熊 2013 物理学报 62 064705]
[3]	Xu S F, Wang J G 2013 Acta Phys. Sin. 62 124701 (in Chinese) [许少锋, 汪久根 2013 物理学报 62 124701]
[4]	Chang J Z, Liu M B 2010 Acta Phys. Sin. 59 7556 (in Chinese) [常建忠, 刘谋斌 2010 物理学报 59 7556]
[5]	Wang X L, Chen S 2010 Acta Phys. Sin. 59 6778 (in Chinese) [王晓亮, 陈硕 2010 物理学报 59 6778]
[6]	Wu S F, Li X F 2007 Chin. Phys. Lett. 25 184
[7]	He L L, Zhang R F, Ji Y Y 2012 Chin. Phys. B 21 088301
[8]	Liu C F, Ni Y S 2008 Chin. Phys. B 17 4554
[9]	Zhong C W, Xie J F, Zhuo C S, Xiong S W, Yin D C 2009 Chin. Phys. B 18 4083
[10]	Li Q D, Tan Y L, Yang F Y 2011 Acta Phys. Sin. 60 030206 (in Chinese) [李清都, 谭宇玲, 杨芳艳 2011 物理学报 60 030206]
[11]	Li Q D, Zhou H W, Yang X S 2012 Acta Phys. Sin. 61 040503 (in Chinese) [李清都, 周红伟, 杨晓松 2012 物理学报 61 040503]
[12]	Wang X C, Yan B, Liu H K, Li L, Wei X, Hu G E 2013 Acta Phys. Sin. 62 098702 (in Chinese) [汪先超, 闫镔, 刘宏奎, 李磊, 魏星, 胡国恩 2013 物理学报 62 098702]
[13]	Huang C S, Zhang W H, Hou Z M, Chen J H, Li M J, He N Z, Shi B C 2011 Chin. Sci. Bull. 56 2829 (in Chinese) [黄昌盛, 张文欢, 侯志敏, 陈俊辉, 李明晶, 何南忠, 施保昌 2011 科学通报 56 2829]
[14]	Zheng Y C, Liu S, Xiong S W, Zhou J F 2010 Sci. Tech. Eng. 7 1684 (in Chinese) [郑彦奎, 刘沙, 熊生伟, 周季夫 2010 科学技术与工程 7 1684]
[15]	Zhang D D, Xu Y, Xu L 2012 Comput. Sci. 39 296 (in Chinese) [张丹丹, 徐莹, 徐磊 2012 计算机科学 39 296]
[16]	Li C G, Maa Jerome P Y, Kang H G 2012 Sci. China: Phys. Mech. Astron. 55 1894
[17]	Januszewski M, Kostur M 2010 Comput. Phys. Commun. 181 183
[18]	Yuen D A, Wang L 2013 GPU Solutions to Multi-scale Problems in Science and Engineering (Berlin: Springer-Verlag) p143
[19]	Wu H, Xu J B, Zhang S F, Wen H 2011 IEIT J. Adapt. Dyn. Comput. 4 26
[20]	Wang S B, Xua J B, Wen H 2013 Comput. Phys. Commun. 184 2454
[21]	Chen S, Jin Y B, Zhang M K, Shang Z 2012 J. Tongji Univ. (Natural Science) 40 137 (in Chinese) [陈硕, 金亚斌, 张明焜, 尚智 2012 同济大学学报 (自然科学版) 40 137]
[22]	Phillips C L, Andersonb J A, Glotzer S C 2011 J. Comput. Phys. 230 7191
[23]	Howes L, Thomas D 2007 Efficient Random Number Generation and Application Using CUDA (Boston: Addison-Wesley Professional) p370
[24]	Nandapalan N, Brent R P, Murray L M, Rendell A 2012 Parallel Processing and Applied Mathematics (Berlin: Springer-Verlag) p609
[25]	Rose G 2011 IACR Cryptology ePrint Archive 2011 7
[26]	Wheeler D J, Needham R M 1995 Fast Software Encryption (Berlin: Springer-Verlag) p363
[27]	Yao P 2010 M. S. Dissertation (Hefei: University of Science and Technology of China) (in Chinese) [姚平 2010 硕士学位论文 (合肥: 中国科学技术大学)]
[28]	Fan X J, Nhan P T, Yong N T, Wu X H, Xu D 2003 Phys. Fluids 15 11

施引文献

[1]	Zhang M K, Chen S, Shang Z 2012 Acta Phys. Sin. 61 034701 (in Chinese) [张明焜, 陈硕, 尚智 2012 物理学报 61 034701]
[2]	Liu H T, Liu M B, Chang J Z, Su T X 2013 Acta Phys. Sin. 62 064705 (in Chinese) [刘汉涛, 刘谋斌, 常建忠, 苏铁熊 2013 物理学报 62 064705]
[3]	Xu S F, Wang J G 2013 Acta Phys. Sin. 62 124701 (in Chinese) [许少锋, 汪久根 2013 物理学报 62 124701]
[4]	Chang J Z, Liu M B 2010 Acta Phys. Sin. 59 7556 (in Chinese) [常建忠, 刘谋斌 2010 物理学报 59 7556]
[5]	Wang X L, Chen S 2010 Acta Phys. Sin. 59 6778 (in Chinese) [王晓亮, 陈硕 2010 物理学报 59 6778]
[6]	Wu S F, Li X F 2007 Chin. Phys. Lett. 25 184
[7]	He L L, Zhang R F, Ji Y Y 2012 Chin. Phys. B 21 088301
[8]	Liu C F, Ni Y S 2008 Chin. Phys. B 17 4554
[9]	Zhong C W, Xie J F, Zhuo C S, Xiong S W, Yin D C 2009 Chin. Phys. B 18 4083
[10]	Li Q D, Tan Y L, Yang F Y 2011 Acta Phys. Sin. 60 030206 (in Chinese) [李清都, 谭宇玲, 杨芳艳 2011 物理学报 60 030206]
[11]	Li Q D, Zhou H W, Yang X S 2012 Acta Phys. Sin. 61 040503 (in Chinese) [李清都, 周红伟, 杨晓松 2012 物理学报 61 040503]
[12]	Wang X C, Yan B, Liu H K, Li L, Wei X, Hu G E 2013 Acta Phys. Sin. 62 098702 (in Chinese) [汪先超, 闫镔, 刘宏奎, 李磊, 魏星, 胡国恩 2013 物理学报 62 098702]
[13]	Huang C S, Zhang W H, Hou Z M, Chen J H, Li M J, He N Z, Shi B C 2011 Chin. Sci. Bull. 56 2829 (in Chinese) [黄昌盛, 张文欢, 侯志敏, 陈俊辉, 李明晶, 何南忠, 施保昌 2011 科学通报 56 2829]
[14]	Zheng Y C, Liu S, Xiong S W, Zhou J F 2010 Sci. Tech. Eng. 7 1684 (in Chinese) [郑彦奎, 刘沙, 熊生伟, 周季夫 2010 科学技术与工程 7 1684]
[15]	Zhang D D, Xu Y, Xu L 2012 Comput. Sci. 39 296 (in Chinese) [张丹丹, 徐莹, 徐磊 2012 计算机科学 39 296]
[16]	Li C G, Maa Jerome P Y, Kang H G 2012 Sci. China: Phys. Mech. Astron. 55 1894
[17]	Januszewski M, Kostur M 2010 Comput. Phys. Commun. 181 183
[18]	Yuen D A, Wang L 2013 GPU Solutions to Multi-scale Problems in Science and Engineering (Berlin: Springer-Verlag) p143
[19]	Wu H, Xu J B, Zhang S F, Wen H 2011 IEIT J. Adapt. Dyn. Comput. 4 26
[20]	Wang S B, Xua J B, Wen H 2013 Comput. Phys. Commun. 184 2454
[21]	Chen S, Jin Y B, Zhang M K, Shang Z 2012 J. Tongji Univ. (Natural Science) 40 137 (in Chinese) [陈硕, 金亚斌, 张明焜, 尚智 2012 同济大学学报 (自然科学版) 40 137]
[22]	Phillips C L, Andersonb J A, Glotzer S C 2011 J. Comput. Phys. 230 7191
[23]	Howes L, Thomas D 2007 Efficient Random Number Generation and Application Using CUDA (Boston: Addison-Wesley Professional) p370
[24]	Nandapalan N, Brent R P, Murray L M, Rendell A 2012 Parallel Processing and Applied Mathematics (Berlin: Springer-Verlag) p609
[25]	Rose G 2011 IACR Cryptology ePrint Archive 2011 7
[26]	Wheeler D J, Needham R M 1995 Fast Software Encryption (Berlin: Springer-Verlag) p363
[27]	Yao P 2010 M. S. Dissertation (Hefei: University of Science and Technology of China) (in Chinese) [姚平 2010 硕士学位论文 (合肥: 中国科学技术大学)]
[28]	Fan X J, Nhan P T, Yong N T, Wu X H, Xu D 2003 Phys. Fluids 15 11

计量

文章访问数: 5977
PDF下载量: 574
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

搜索

留言板