期刊文献+

分子动力学并行算法的优化与应用 被引量:6

Optimization and application of parallel molecular dynamics simulation
下载PDF
导出
摘要 比较了小型机群中模拟小规模粒子系统采用原子分解和区域分解算法的模拟时间。针对原子分解算法,优化了各进程粒子的受力计算,尤其对严重影响并行效率的全局通信进行了优化,最终使得并行效率提高了30%以上。使用该优化算法对固态氩的导热系数进行了计算,模拟结果与实验值比较吻合,大大缩短了时间。 This paper compared the simulation time of adopting atom-decomposition algorithm and domain-decomposition algo- rithm when simulating small-scale atom system in small-scale cluster. According to the atom-decomposition algorithm, optimized the force calculation of every processer and global communication which affected the parallel efficiency seriously, And it would enhance the parallel efficiency to more than 30%, At last calculated the thermal conductivity of solid argon by the use of the optimized algorithm, The simulation results are very consistent with the experimental ones, and the simulation time has reduced much.
出处 《计算机应用研究》 CSCD 北大核心 2008年第3期718-720,共3页 Application Research of Computers
基金 国家自然科学基金资助项目(50506008,50475077)
关键词 分子动力学 并行算法 原子分解法 MPI_Allgather 邻居交换算法 非阻塞通信 molecular dynamics parallel algorithm atom decomposition MPI_Allgather neighbor-exchange algorithm nonblocking communication
  • 相关文献

参考文献8

  • 1ALLEN M P, TILDESLEY D J. Computer simulation of liquids [ M ], Oxford: Clarendon Press, 1987.
  • 2PLIMPTON S. Fast parallel algorithms for short-range molecular dynamics [J]. Journal of Computational Physics, 1995,117 ( 1 ) : 1 - 19.
  • 3吴江涛,刘志刚,董小社,何戈.RS/6000机群系统中分子动力学并行算法的研究[J].工程热物理学报,2002,23(4):418-420. 被引量:3
  • 4BROWN D,CLARKE J H R,OKUDA M,et aL A domain decomposition parallelization strategy for molecular dynamics simulations on distributed memory machines [ J ]. Comput Phy Commun, 1993,74 ( 1 ) :67-80.
  • 5BRUCK J, HO C, KIPNIS S, et al. Efficient algorithms for all-to-all communications in multiport message passing systems [ J ], IEEE Trans on Parallel and Distributed Systems, 1997,8 ( 11 ) : 1143- 1156.
  • 6RAJEEV T,WILLIAM G. Improving the performance of MPI collective communication on switched networks[ C ]//Proc of the 10th European PVM/MPI User's Group Meeting. Venice: [ s. n. ] ,2003:257-267.
  • 7陈靖,张云泉,张林波,袁伟.一种新的MPI Allgather算法及其在万亿次机群系统上的实现与性能分析[J].计算机学报,2006,29(5):808-814. 被引量:8
  • 8CHRISTENT D K, POLLACK G L. Thermal conductivity of solid argon [J]. Phys Rev B,1975,12(8) :3380-3391.

二级参考文献8

  • 1Benson G.D,Chu C.W.et al.A comparison of MPICH Allgather algorithms on switched networks.In:Proceedings of the 10th European PVM/MPI Users' Group Meeting,Venice,2003,335~343
  • 2Hensgen D,Finkel R,Manbet U..Two algorithms for barrier synchronization.International Journal of Parallel Programming,1988,17(1):1~17
  • 3Bruck J,Ho C.T,Kipnis S.et al.Efficient algorithms for all-to-all communications in multiport message passing systems.IEEE Transactions on Parallel and Distributed Systems,1997,8(11):1143~1156
  • 4Chan E.W,Heimlich M.F,Purakayastha A.et al.On optimizing collective communication.In:Proceedings of the 2004 IEEE International Conference on Cluster Computing,San Diego,2004,145~155
  • 5Thakur R,Rabenseifner R.et al.Optimization of collective communication operations in MPICH.International Journal of High Performance Computing Applications,2005,19(1):49~66
  • 6Johnson K.L..The impact of communication locality on large-scale multiprocessor performance.ACM SIGARCH Computer Architecture News,1992,20(2):392~402
  • 7http//www-unix.mcs.anl.gov/mpi/mpich/.
  • 8吴江涛,刘志刚,赵小明.分子动力学模拟中不同短程作用力计算方法的效率研究[J].西安交通大学学报,2002,36(5):477-481. 被引量:3

共引文献9

同被引文献40

引证文献6

二级引证文献14

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部