Parallel implementation of 3D FFT with volumetric decomposition schemes for efficient molecular dynamics simulations
Parallel implementation of 3D FFT with volumetric decomposition schemes for efficient molecular dynamics simulations
复制标题
DOI：
10.1016/j.cpc.2015.10.024
复制
发表时间：
2016-03-01
影响因子：
6.3
通讯作者：
Sugita, Yuji
中科院分区：
文献类型：
--
作者：
Jung, Jaewoon;Kobayashi, Chigusa;Sugita, Yuji
三维快速傅立叶变换(3DFFT)在包括分子动力学(MD)模拟在内的各种计算机模拟和数据分析中扮演着重要的角色。在这项研究中，我们开发了基于两种新的体积分解的混合(MPI+OpenMP)3D FFT并行化方案，主要用于MD模拟中的粒子网格Ewald(PME)计算。在一个方案(1D_ALLTOALL)中，在一个维度上执行五个全对所有通信，而在另一个方案(2D_ALLTOALL)中，一个二维全到所有通信与一个维度上的两个全到所有通信相结合。2D_Alltoall类似于传统的体积分解方案。我们在RIKEN AICS的K计算机上使用大量处理器对不同网格大小的系统进行了3DFFT的基准测试。这两种方案的性能相当，且优于现有的3DFFT。1D_ALLTOALL和2D_ALLTOALL的性能取决于超级计算机网络系统和每个维度上的处理器数量。用户有足够的回旋余地来根据自己的条件优化性能。在PME方法中，既计算了短程实空间相互作用，也计算了长程互易空间相互作用。我们的体积分解方案在与最近开发的用于短程相互作用的中点单胞方法结合使用时特别有用，因为真实空间和倒易空间的分解是相同的。3DFFT的1D_ALLITOALL格式在K计算机上用32,768个核模拟了一个包含100多万个原子的病毒系统的一个MD周期，耗时4.7ms。(C)2015爱思唯尔B.V.保留所有权利。
Three-dimensional Fast Fourier Transform (3D FFT) plays an important role in a wide variety of computer simulations and data analyses, including molecular dynamics (MD) simulations. In this study, we develop hybrid (MPI+OpenMP) parallelization schemes of 3D FFT based on two new volumetric decompositions, mainly for the particle mesh Ewald (PME) calculation in MD simulations. In one scheme, (1d_Alltoall), five all-to-all communications in one dimension are carried out, and in the other, (2d_Alltoall), one two-dimensional all-to-all communication is combined with two all-to-all communications in one dimension. 2d_Alltoall is similar to the conventional volumetric decomposition scheme. We performed benchmark tests of 3D FFT for the systems with different grid sizes using a large number of processors on the K computer in RIKEN AICS. The two schemes show comparable performances, and are better than existing 3D FFTs. The performances of 1d_Alltoall and 2d_Alltoall depend on the supercomputer network system and number of processors in each dimension. There is enough leeway for users to optimize performance for their conditions. In the PME method, short-range real-space interactions as well as long-range reciprocal-space interactions are calculated. Our volumetric decomposition schemes are particularly useful when used in conjunction with the recently developed midpoint cell method for short-range interactions, due to the same decompositions of real and reciprocal spaces. The 1d_Allitoall scheme of 3D FFT takes 4.7 ms to simulate one MD cycle for a virus system containing more than 1 million atoms using 32,768 cores on the K computer. (C) 2015 Elsevier B.V. All rights reserved.