PuReMD-GPU: A reactive molecular dynamics simulation package for GPUs

被引:43
作者
Kylasa, S. B. [1 ]
Aktulga, H. M. [3 ]
Grama, A. Y. [2 ]
机构
[1] Purdue Univ, Dept Elec & Comp Eng, W Lafayette, IN 47907 USA
[2] Purdue Univ, Dept Comp Sci, W Lafayette, IN 47907 USA
[3] Univ Calif Berkeley, Lawrence Berkeley Natl Lab, Berkeley, CA 94720 USA
关键词
GPUs; GPU-enabled applications; Reactive molecular dynamics; Performance optimizations; FORCE-FIELD; REAXFF; HYDROCARBONS; ALGORITHMS;
D O I
10.1016/j.jcp.2014.04.035
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
We present an efficient and highly accurate GP-GPU implementation of our community code, PuReMD, for reactive molecular dynamics simulations using the ReaxFF force field. PuReMD and its incorporation into LAMMPS (Reax/C) is used by a large number of research groups worldwide for simulating diverse systems ranging from biomembranes to explosives (RDX) at atomistic level of detail. The sub-femtosecond time-steps associated with ReaxFF strongly motivate significant improvements to per-timestep simulation time through effective use of GPUs. This paper presents, in detail, the design and implementation of PuReMD-GPU, which enables ReaxFF simulations on GPUs, as well as various performance optimization techniques we developed to obtain high performance on state-of-the-art hardware. Comprehensive experiments on model systems (bulk water and amorphous silica) are presented to quantify the performance improvements achieved by PuReMD-GPU and to verify its accuracy. In particular, our experiments show up to 16 x improvement in runtime compared to our highly optimized CPU-only single-core ReaxFF implementation. PuReMD-GPU is a unique production code, and is currently available on request from the authors. (C) 2014 Elsevier Inc. All rights reserved.
引用
收藏
页码:343 / 359
页数:17
相关论文
共 33 条
[1]   Parallel reactive molecular dynamics: Numerical methods and algorithmic techniques [J].
Aktulga, H. M. ;
Fogarty, J. C. ;
Pandit, S. A. ;
Grama, A. Y. .
PARALLEL COMPUTING, 2012, 38 (4-5) :245-259
[2]  
Aktulga H. M., LAMMPS USER REAX C
[3]  
Aktulga H.M., 2010, THESIS PURDUE U WEST
[4]  
Aktulga H.M., 2010, CSRI SUMM P 2009, P50
[5]   REACTIVE MOLECULAR DYNAMICS: NUMERICAL METHODS AND ALGORITHMIC TECHNIQUES [J].
Aktulga, Hasan Metin ;
Pandit, Sagar A. ;
van Duin, Adri C. T. ;
Grama, Ananth Y. .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 2012, 34 (01) :C1-C23
[6]  
[Anonymous], CUDA C Best Practices Guide
[7]  
[Anonymous], 2003, ITERATIVE METHODS SP, DOI DOI 10.1137/1.9780898718003
[8]  
Bell N., 2008, Efficient sparse matrix-vector multiplication on CUDA
[9]   EMPIRICAL POTENTIAL FOR HYDROCARBONS FOR USE IN SIMULATING THE CHEMICAL VAPOR-DEPOSITION OF DIAMOND FILMS [J].
BRENNER, DW .
PHYSICAL REVIEW B, 1990, 42 (15) :9458-9471
[10]   A second-generation reactive empirical bond order (REBO) potential energy expression for hydrocarbons [J].
Brenner, DW ;
Shenderova, OA ;
Harrison, JA ;
Stuart, SJ ;
Ni, B ;
Sinnott, SB .
JOURNAL OF PHYSICS-CONDENSED MATTER, 2002, 14 (04) :783-802