MDSLB：A new static load balancing method for parallel molecular dynamics simulations

doi:10.1088/1674-1056/23/2/028903

Abstract Large-scale parallelization of molecular dynamics simulations is facing challenges which seriously affect the simulation efficiency, among which the load imbalance problem is the most critical. In this paper, we propose, a new molecular dynamics static load balancing method (MDSLB). By analyzing the characteristics of the short-range force of molecular dynamics programs running in parallel, we divide the short-range force into three kinds of force models, and then package the computations of each force model into many tiny computational units called “cell loads”, which provide the basic data structures for our load balancing method. In MDSLB, the spatial region is separated into sub-regions called “local domains”, and the cell loads of each local domain are allocated to every processor in turn. Compared with the dynamic load balancing method, MDSLB can guarantee load balance by executing the algorithm only once at program startup without migrating the loads dynamically. We implement MDSLB in OpenFOAM software and test it on TianHe-1A supercomputer with 16 to 512 processors. Experimental results show that MDSLB can save 34%–64% time for the load imbalanced cases.

Keywords: molecular dynamics static load balancing parallel computing

Received: 30 May 2013
Revised: 04 September 2013
Accepted manuscript online:

PACS:	89.20.-a	(Interdisciplinary applications of physics)
	89.20.Ff	(Computer science and technology)
	02.70.Ns	(Molecular dynamics and particle methods)
	31.15.-p	(Calculations and mathematical techniques in atomic and molecular physics)

Fund: Project supported by the National Natural Science Foundation of China (Grant Nos. 61303071 and 61120106005) and the Natural Science Fund from the Guangzhou Science and Information Technology Bureau (Grant No. 134200026).

Corresponding Authors: Xu Xin-Hai
E-mail: xuxinhai@nudt.edu.cn

About author: 89.20.-a; 89.20.Ff; 02.70.Ns; 31.15.-p

Cite this article:

Wu Yun-Long (武云龙), Xu Xin-Hai (徐新海), Yang Xue-Jun (杨学军), Zou Shun (邹顺), Ren Xiao-Guang (任小广) MDSLB：A new static load balancing method for parallel molecular dynamics simulations 2014 Chin. Phys. B 23 028903

[1]	Rahman A 1964 Phys. Rev. 136 405
[2]	Zhu R Z and Yan H 2011 Chin. Phys. B 20 016801
[3]	Harvey M J and De Fabritiis G 2012 Drug Discovery Today 17 1059
[4]	Wu T Y, Lai W S and Fu B Q 2013 Chin. Phys. B 22 076601
[5]	Chai A H, Ran S Y, Zhang D, Jiang Y W, Yang G C and Zhang L X 2013 Chin. Phys. B 22 098701
[6]	Maginn E J and Elliott J R 2010 Ind. Eng. Chem. Res. 49 3059
[7]	Imran M, Hussain F, Rashid M and Ahmad S A 2012 Chin. Phys. B 21 116201
[8]	Mo Z Y, Zhang J L and Cai Q D 2002 Int. J. Comput. Math. 79 165
[9]	Rapaport D C 2004 The Art of Molecular Dynamics Simulation (New York: Cambridge University Press)
[10]	Kumar S, Huang C, Zheng, G, Bohm E, Bhatelé A, Phillips J C, Yu H and Kalé L V 2008 IBM J. Res. & Dev. 52 177
[11]	Zhu X Q, Liu X, Jian X D, Meng X F and Feng J H 2012 In Advances in Information Technology and Industry Applications (Berlin: Springer-Verlag) p. 265
[12]	Michael B W, Nguyen T D, Fuentes-Cabrera M, Fowlkes J D, Rack P D, Berger M and Bland A S 2012 Proceedings of the International Conference on Computational Science, June 4–6, 2012, Omaha, USA, p. 186
[13]	Bland B, Wells J, Messer B, Hernandez O and R Jim 2012 https://cug.org/proceedings/attendee_program_cug2012/includes/files/pap138file2.pdf [2012-05-02]
[14]	Alam S R, Vetter J S, Agarwal P K and Geist A 2006 Proceedings of the ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, March 29–31 2006, New York, USA, p. 59
[15]	Bhatelé A, Kalé L V and Kumar S 2009 Proceedings of the 23rd International Conference on Supercomputing, June 8–12, 2009, Yorktown Heights, USA, p. 110
[16]	Deng Y F, Peierlsc R F and Riverad C 2000 J. Comput. Phys. 161 250
[17]	Kalé L V, Skeel R, Bhandarkar M, Brunner R, Gursoy A, Krawetz N, Phillips J, Shinozaki A, Varadarajan K and Schulten K 1999 J. Comput. Phys. 151 283
[18]	Brown W M, Wang P, Plimpton S J and Tharrington A N 2011 Comput. Phys. Commun. 182 898
[19]	Allen M P and Tildesley D J 1989 Computer Simulation of Liquids (Oxford: Clarendon Press)
[20]	Verlet L 1967 Phys. Rev. 159 98
[21]	Quentrec B and Brot C 1973 J. Comput. Phys. 13 430
[22]	Hockney R W and Eastwood J W 1988 Computer Simulation Using Particles (Bristol: Taylor & Francis)
[23]	Plimpton S 1995 J. Comput. Phys. 117 1
[24]	Chandy J A, Kim S, Ramkumar B, Parkes S and Banerjee P 1997 IEEE Trans. Comput. Aided Des. Integr. Circuits Syst. 16 398
[25]	Nour-Omid B, Raefsky A and Lyzenga G 1987 Proceedings of the Symposium on Parallel Computations and Their Impact on Mechanics, December 13–18, 1987, Boston, USA, p. 209
[26]	Bially T 1969 IEEE Trans. Inf. Theory 15 658
[27]	Karypis G and Kumar V 1999 Proceedings of the 36th Annual ACM/IEEE Design Automation Conference, June 21–25, 1999, New Orleans, USA, p. 343
[28]	Karypis G, Aggarwal R, Kumar V and Shekhar S 1997 Proceedings of the 34th Annual Design Automation Conference, June 9–13, 1997, Anaheim, USA, p. 526
[29]	Kikuchi H, Karki B B and Saini S 2006 Proceedings of the International Conference on Parallel and Distributed Processing Techniques and Applications, June 26–29, 2006, Las Vegas, USA, p. 1083
[30]	Hess B, Kutzner C, van der Spoel D and Lindahl E 2008 J. Chem. Theory Comput. 4 435
[31]	Larsson P, Hess B and Lindahl E 2011 Wiley Interdisciplin. Rev. Comput. Mol. Sci. 1 93
[32]	Hayashi R and Horiguchi S 2000 Proceedings of the 14th International Parallel & Distributed Processing Symposium, May 1–5, 2000, Cancun, Mexico, p. 85
[33]	Zhakhovskii V, Nishihara K, Fukuda Y, Shimojo S, Akiyama T, Miyanaga S, Sone H, Kobayashi H, Ito E, Seo Y, Tamura M and Ueshima Y 2005 Proceedings of the 5th International Symposium on Cluster Computing and the Grid, May 9–12, 2005, Cardiff, UK, p. 848
[34]	Fattebert J L, Richards D F and Glosli J N 2012 Comput. Phys. Commun. 183 2608

[1]	Molecular dynamics study of interactions between edge dislocation and irradiation-induced defects in Fe–10Ni–20Cr alloy Tao-Wen Xiong(熊涛文), Xiao-Ping Chen(陈小平), Ye-Ping Lin(林也平), Xin-Fu He(贺新福), Wen Yang(杨文), Wang-Yu Hu(胡望宇), Fei Gao(高飞), and Hui-Qiu Deng(邓辉球). Chin. Phys. B, 2023, 32(2): 020206.
[2]	Formation of nanobubbles generated by hydrate decomposition: A molecular dynamics study Zilin Wang(王梓霖), Liang Yang(杨亮), Changsheng Liu(刘长生), and Shiwei Lin(林仕伟). Chin. Phys. B, 2023, 32(2): 023101.
[3]	Prediction of flexoelectricity in BaTiO₃ using molecular dynamics simulations Long Zhou(周龙), Xu-Long Zhang(张旭龙), Yu-Ying Cao(曹玉莹), Fu Zheng(郑富), Hua Gao(高华), Hong-Fei Liu(刘红飞), and Zhi Ma(马治). Chin. Phys. B, 2023, 32(1): 017701.
[4]	Adsorption dynamics of double-stranded DNA on a graphene oxide surface with both large unoxidized and oxidized regions Mengjiao Wu(吴梦娇), Huishu Ma(马慧姝), Haiping Fang(方海平), Li Yang(阳丽), and Xiaoling Lei(雷晓玲). Chin. Phys. B, 2023, 32(1): 018701.
[5]	Effect of spatial heterogeneity on level of rejuvenation in Ni₈₀P₂₀ metallic glass Tzu-Chia Chen, Mahyuddin KM Nasution, Abdullah Hasan Jabbar, Sarah Jawad Shoja, Waluyo Adi Siswanto, Sigiet Haryo Pranoto, Dmitry Bokov, Rustem Magizov, Yasser Fakri Mustafa, A. Surendar, Rustem Zalilov, Alexandr Sviderskiy, Alla Vorobeva, Dmitry Vorobyev, and Ahmed Alkhayyat. Chin. Phys. B, 2022, 31(9): 096401.
[6]	Spatial correlation of irreversible displacement in oscillatory-sheared metallic glasses Shiheng Cui(崔世恒), Huashan Liu(刘华山), and Hailong Peng(彭海龙). Chin. Phys. B, 2022, 31(8): 086108.
[7]	Effect of void size and Mg contents on plastic deformation behaviors of Al-Mg alloy with pre-existing void: Molecular dynamics study Ning Wei(魏宁), Ai-Qiang Shi(史爱强), Zhi-Hui Li(李志辉), Bing-Xian Ou(区炳显), Si-Han Zhao(赵思涵), and Jun-Hua Zhao(赵军华). Chin. Phys. B, 2022, 31(6): 066203.
[8]	Strengthening and softening in gradient nanotwinned FCC metallic multilayers Yuanyuan Tian(田圆圆), Gangjie Luo(罗港杰), Qihong Fang(方棋洪), Jia Li(李甲), and Jing Peng(彭静). Chin. Phys. B, 2022, 31(6): 066204.
[9]	Investigation of the structural and dynamic basis of kinesin dissociation from microtubule by atomistic molecular dynamics simulations Jian-Gang Wang(王建港), Xiao-Xuan Shi(史晓璇), Yu-Ru Liu(刘玉如), Peng-Ye Wang(王鹏业),Hong Chen(陈洪), and Ping Xie(谢平). Chin. Phys. B, 2022, 31(5): 058702.
[10]	Impact of thermostat on interfacial thermal conductance prediction from non-equilibrium molecular dynamics simulations Song Hu(胡松), C Y Zhao(赵长颖), and Xiaokun Gu(顾骁坤). Chin. Phys. B, 2022, 31(5): 056301.
[11]	Evaluation on performance of MM/PBSA in nucleic acid-protein systems Yuan-Qiang Chen(陈远强), Yan-Jing Sheng(盛艳静), Hong-Ming Ding(丁泓铭), and Yu-Qiang Ma(马余强). Chin. Phys. B, 2022, 31(4): 048701.
[12]	Molecular dynamics simulations of A-DNA in bivalent metal ions salt solution Jingjing Xue(薛晶晶), Xinpeng Li(李新朋), Rongri Tan(谈荣日), and Wenjun Zong(宗文军). Chin. Phys. B, 2022, 31(4): 048702.
[13]	Evolution of defects and deformation mechanisms in different tensile directions of solidified lamellar Ti-Al alloy Yutao Liu(刘玉涛), Tinghong Gao(高廷红), Yue Gao(高越), Lianxin Li(李连欣), Min Tan(谭敏), Quan Xie(谢泉), Qian Chen(陈茜), Zean Tian(田泽安), Yongchao Liang(梁永超), and Bei Wang(王蓓). Chin. Phys. B, 2022, 31(4): 046105.
[14]	Effect of the number of defect particles on the structure and dispersion relation of a two-dimensional dust lattice system Rangyue Zhang(张壤月), Guannan Shi(史冠男), Hanyu Tang(唐瀚宇), Yang Liu(刘阳), Yanhong Liu(刘艳红), and Feng Huang(黄峰). Chin. Phys. B, 2022, 31(3): 035204.
[15]	Molecular dynamics simulations on the wet/dry self-latching and electric fields triggered wet/dry transitions between nanosheets: A non-volatile memory nanostructure Jianzhuo Zhu(朱键卓), Xinyu Zhang(张鑫宇), Xingyuan Li(李兴元), and Qiuming Peng(彭秋明). Chin. Phys. B, 2022, 31(2): 024703.

No Suggested Reading articles found!

Viewed

Full text

Abstract

Cited

Metrics
Related Articles

MDSLB：A new static load balancing method for parallel molecular dynamics simulations

Cite this article:

Online attention