基于弹性云计算集群的国际TEAM Problem 7基准问题计算方法

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (16997 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract Different development models of cloud computing were comparatively analyzed for intelligent manufacturing in this paper, focusing on large-scale computing problems. A kind of compute-intensive cloud architecture has been chosen, and then an elastic computing platform containing nine computing nodes has been set up. Taken the international benchmark problem of TEAM 7 in the electromagnetic field as a case, the parallel computing program of the typical eddy current field was written by the open source parallel programming model OpenMpi, based on the message passing interface. The parallel computing cases, with hundreds of large-scale computing nodes, have been carried out. Accordingly, the distribution graphs of computational time and rate under different computational scales and different nodes were painted. It is proved that the parallelization of the large-scale finite element simulation of typical electrical equipment can be achieved in the elastic computing cluster, which is deployed in the cloud-computing platform. Compared with the traditional high-performance parallel computing method, the elastic computing cluster in cloud computing has a series of merits, such as low investing cost, dynamic distribution, obtain on demand. Thus a flexible and convenient new computing scheme is provided for upgrading from pure traditional manufacturing industry to intelligent manufacturing industry impelled by innovation.

Key words： Cloud computing, high performance computing, eddy current field, finite element method

Received: 03 May 2016 Published: 02 May 2017

PACS:

TM152

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Jin Liang
	Wang Dongmei
	Qiu Yuntao
	Yang Qingxin
	Niu Pingjuan

Cite this article:

Jin Liang,Wang Dongmei,Qiu Yuntao等. A Calculation Method for the International Benchmark Problem of TEAM 7 Based on Elastic Cluster in Cloud Computing[J]. Transactions of China Electrotechnical Society, 2017, 32(8): 144-150.

URL:

https://dgjsxb.ces-transaction.com/EN/ OR https://dgjsxb.ces-transaction.com/EN/Y2017/V32/I8/144

[1] 陈国良, 毛睿, 蔡晔. 高性能计算及其相关新兴技术[J]. 深圳大学学报(理工版), 2015, 32(1): 25-31.
Chen Guoliang, Mao Rui, Cai Ye. High performance computing and related new technologies[J]. Journal of Shenzhen University (Science Engineering), 2015, 32(1): 25-31.
[2] 王普勇, 丁峻宏. 高性能计算在工业工程领域的应用[J]. 科研信息化技术与应用, 2012, 3(6): 3-11.
Wang Puyong, Ding Junhong. Application and practice of high performance computing in industry and engineering[J]. E-Science Technology & Appli- cation, 2012, 3(6): 3-11.
[3] 龚春叶, 包为民, 汤国建, 等. 航天领域高性能并行计算研究进展[J]. 计算机工程与科学, 2014, 36(9): 1629-1636.
Gong Chunye, Bao Weimin, Tang Guojian, et al. Recent progress in high-performance parallel com- puting of the aerospace area[J]. Computer Engin- eering and Science, 2014, 36(9): 1629-1636.
[4] 周兴铭. 高性能计算技术发展[J]. 自然杂志, 2011, 33(5): 249-254.
Zhou Xingming. High performance computing[J]. Chinese Journal of Nature, 2011, 33(5): 249-254.
[5] 杜永. 面向变压器工程的铁磁材料电磁性能的模拟研究[D]. 天津: 河北工业大学, 2010.
[6] 商磊, 吕全义. 并行技术在电磁有限元计算中的应用[J]. 西南民族大学学报(自然科学版), 2007, 33(2): 291-294.
Shang Lei, Lü Quanyi. Application of parallel technology to electromagnetic FEM computation[J]. Journal of Southwest University for Nationalities (Natural Science Edition), 2007, 33(2): 291-294.
[7] 厉天威, 阮江军, 黄道春, 等. 大规模电磁场数值计算中并行迭代方法的比较[J]. 电工技术学报, 2007, 22(8): 166-173.
Li Tianwei, Ruan Jiangjun, Huang Daochun, et al. Comparison of parallel iterative methods on large- scale electromagnetic field numerical computation[J]. Transactions of China Electrotechnical Society, 2007, 22(8): 166-173.
[8] Jorissen K, Vila Fd, Rehr J J. A high performance scientific cloud computing environment for materials simulations[J]. Computer Physics Communications, 2012, 183(9): 1911-1919.
[9] Lin Fengsheng, Shen Chiaping, Liu Chiahung, et al. High-performance multiclass classification frame- work using cloud computing architecture[J]. Journal of Medical and Biological Engineering, 2015, 35(6): 795-802.
[10] Li H, Lu J. Simulation and optimization of electromagnetic compatibility problems using com- putational electromagnetics and high performance computing: a cloud computing based system[C]// Sixth International Conference on Electromagnetic Field Problems and Applications (ICEF), 2012: 1-4.
[11] 赵俊华, 文福拴, 薛禹胜, 等. 云计算: 构建未来电力系统的核心计算平台[J]. 电力系统自动化, 2010, 34(15): 1-8.
Zhao Junhua, Wen Fushuan, Xue Yusheng, et al. Cloud computing: implementing an essential com- puting platform for future power systems[J]. Auto- mation of Electric Power Systems, 2010, 34(15): 1-8.
[12] Golpayegani N, Halem M. Cloud computing for satellite data processing on high end compute clusters[C]//Proceedings of 2009 IEEE International Conference on Cloud Computing, 2009: 88-92.
[13] Mauch Viktor, Kunze Marcel, Hillenbrand Marius. High performance cloud computing[J]. Future Gener- ation Computer Systems, 2013, 29(6): 1408-1416.
[14] 杨柳, 张志明. 超算中心和云计算中心融合建设框架研究[J]. 信息系统工程, 2015(1): 93-96.
Yang Liu, Zhang Zhiming. Study on construction of merging architecture of supercomputer center and cloud computing center[J]. Information Systems Engineering, 2015(1): 93-96.
[15] 曾龙海, 张博锋, 张丽华, 等. 基于云计算平台的虚拟集群构建技术研究[J]. 微电子学与计算机, 2010, 27(8): 31-35, 40.
Zeng Longhai, Zhang Bofeng, Zhang Lihua, et al. Virtual cluster constructing based on cloud com- puting platform[J]. Microelectronics & Computer, 2010, 27(8): 31-35, 40.
[16] 郑文武, 李先绪, 黄执勤. 云计算中的并行计算技术分析[J]. 电信科学, 2011, 27(12): 31-38.
Zheng Wenwu, Li Xianxu, Huang Zhiqin. Techno- lical analysis of parallel computing in cloud com- puting[J]. Telecommunications Science, 2011, 27(12): 31-38.
[17] 黄东启, 艾芊, 吴俊宏. 基于MPI并行计算的大规模电力系统电压稳定极限求解[J]. 华东电力, 2012, 40(7): 1148-1152.
Huang Dongqi, Ai Qian, Wu Junhong. Voltage stability limit solution for large-scale power system with parallel computation based on MPI[J]. East China Electric Power, 2012, 40(7): 1148-1152.
[18] 李海江. 基于MPI的并行有限元计算集群的构建[J]. 数值计算与计算机应用, 2004, 25(3): 198-209.
Li Haijiang. MPI based PC cluster distributed finite element analysis[J]. Journal on Numerical Methods and Computer Applications, 2004, 25(3): 198-209.
[19] 王玲. 基于有限元法的第21基准问题求解[J]. 中国新技术新产品, 2010(11): 18.
Wang Ling. A calculation method for the inter- national benchmark problem of TEAM 21 based on FEM[J]. China New Technologies and Products, 2010(11): 18.
[20] 陈兴乐, 雷银照. 铁磁平板脉冲涡流场感应电压时域近似式[J]. 电工技术学报, 2015, 30(10): 14-19.
Chen Xingle, Lei Yinzhao. Time-domain approxi- mate expressions for voltage induced by pulsed eddy current field of ferromagnetic plate[J]. Transactions of China Electrotechnical Society, 2015, 30(10): 14-19.
[21] 唐任远, 吴东阳, 谢德馨. 单元级别并行有限元法求解工程涡流场的关键问题研究[J]. 电工技术学报, 2014, 29(5): 1-8.
Tang Renyuan, Wu Dongyang, Xie Dexin. Research on the key problem of element by element parallel FEM applied to engineering eddy current analysis[J]. Transactions of China Electrotechnical Society, 2014, 29(5): 1-8.
[22] 王兵. 云计算发展及关键技术探讨[J]. 电子测试, 2015(11): 44-45, 41.
Wang Bing. Development and key technologies of cloud computing[J]. Electronic Test, 2015(11): 44-45, 41.
[23] 方巍, 文学志, 潘吴斌, 等. 云计算: 概念、技术及应用研究综述[J]. 南京信息工程大学学报(自然科学版), 2012, 4(4): 351-361.
Fang Wei, Wen Xuezhi, Pan Wubin, et al. Cloud computing: conceptions, key technologies and application[J]. Journal of Nanjing University of Information Science & Technology (Natural Science Edition), 2012, 4(4): 351-361.
[24] Montella Raffaele, Coviello Giuseppe, Giunta Giulio, et al. A general-purpose virtualization service for HPC on cloud computing: an application to GPUs[C]// Proceedings of the 9th International Conference on Parallel Processing and Applied Mathematics, 2012: 740-749.
[25] 张洋. 面向多核处理系统的OpenMPI通信机制研究与优化实现[D]. 长沙: 国防科学技术大学, 2010.
[26] Woodall T S, Graham R L, Castain R H, et al. TEG: a high-performance, scalable, multi-network point- to-point communications methodology[J]. Lecture Notes in Computer Science, 2004, 3241: 303-310.
[27] Hu Yaohui, Xiao Limin, Li Ruan, et al. A network extend method of OpenMPI[C]//IEEE International Conference on Computer Science Information and Technology, 2010: 366-370.