分子动力学并行算法的优化与应用  

Optimization and application of parallel molecular dynamics simulation

作　　者：刘正华[1] 杨决宽[1] 陈云飞[1] 

机构地区：[1]东南大学机械工程学院,南京210096

出　　处：《计算机应用研究》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)

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.

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 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术] TP301.6