基于“块-单元”数据结构的分子动力学并行计算  

Parallel Computation for Molecular Dynamics Simulation Based on Cell-Block Data Structures

作　　者：曹小林[1] 莫则尧[1] 张景琳[1] 陈其峰[1] 

机构地区：[1]北京应用物理与计算数学研究所高性能计算中心计算物理实验室,北京100088

出　　处：《计算物理》2004年第5期377-385,共9页Chinese Journal of Computational Physics

基　　金：国家自然科学基金(60273030);中国工程物理研究院重点基金(2002Z01041)资助项目.

摘　　要：　开发了一种基于"块-单元"数据结构的可扩展并行算法,以实现大规模、非均匀分子动力学模拟.它采用空间填充曲线将三维区域分解转换为一维负载平衡问题,然后用基于实测的多层均权法求解,以保持处理机间负载均衡.在一个MPP并行机的500个CPU上,模拟包含2 1×108个粒子的三维金属微喷射模型,该算法获得了420倍的加速比.A scalable parallel algorithm especially for large-scale three dimensional simulations with seriously non-uniform molecular distributions is presented. In particular, based on cell-block data structures, this algorithm uses space filling curve to convert three-dimensional domain decomposition for load distribution across processors into one-dimensional load balancing problems for which run-time measurement-based multilevel averaging weights method can be applied successfully. Parallel numerical experiments for simulations of three-dimensional metal-eject physical models with 210 million particles using 500 processors of one massively parallel processors have shown that this algorithm has achieved speedup about 420.

关 键 词：数据结构 可扩展 并行计算 CPU 负载均衡 加速比 并行算法 一维 分子动力学模拟 求解 

分 类 号：O241[理学—计算数学] TP393[理学—数学]