A thread-level parallelization of pairwise additive potential and force calculations suitable for current many-core architectures
In molecular dynamics (MD) simulations, calculations of potentials and their derivatives by coordinate, i.e., forces, in a pairwise additive manner such as the Lennard–Jones interactions and a short-range part of the Coulombic interactions form the main part of arithmetic operations. It is essential to achieve high thread-level parallelization efficiency of these pairwise additive calculations of potentials and forces to use current supercomputers with many-core architectures effectively. In this paper, we propose four new thread-level parallelization algorithms for the pairwise additive potential and force calculations. We implement the four codes in a MD calculation code based on the fast multipole method. Performance benchmarks were taken on the FX100 supercomputer and Intel Xeon Phi coprocessor. The code succeeds in achieving high thread-level parallelization efficiency with 32 threads on the FX100 and up to 60 threads on the Xeon Phi.
KeywordsThread-level parallelization Pairwise additive calculations Domain decomposition Molecular dynamics simulation Fast multipole method
We thank Dr. Y. Komura for valuable suggestions to code 1. This work is supported by “Joint Usage/Research Center for Interdisciplinary Large-Scale Information Infrastructures” and “High Performance Computing Infrastructure” in Japan (Project ID jh150015-NA11, jh160040-NAJ, and jh170024-NAH). This work is also supported by the FLAGSHIP2020, MEXT within the priority study5:Development of new fundamental technologies for high-efficiency energy creation, conversion/storage and use (Proposal No. hp170241). This work is partially funded by MEXT’s program for the Development and Improvement for the Next Generation Ultra High-Speed Computer System, under its Subsidies for Operating the Specific Advanced Large Research Facilities (S. S.). Benchmark calculations were taken at the Information Technology Center (ITC) of Nagoya University, and at the ITC of The University of Tokyo. This work is also supported by JSPS KAKENHI Grant Number 16K21094 (Y. A.), and by MEXT KAKENHI Grant No. 26410012 (N. Y.).
- 8.Andoh Y, Yoshii N, Fujimoto K, Mizutani K, Kojima H, Yamada A, Okazaki S, Kawaguchi K, Nagao H, Iwahashi K, Mizutani F, Minami K, Ichikawa S, Komatsu H, Ishizuki S, Takeda Y, Fukushima M (2013) MODYLAS: a highly parallelized general-purpose molecular dynamics simulation program for large-scale systems with long range forces calculated by fast multipole method (FMM) and highly scalable fine-grained new parallel processing algorithms. J Chem Theory Comput 9:3201–3209CrossRefGoogle Scholar
- 9.Harrod WJ (2012) A journey to exascale computing. In: The International Conference for High Performance Computing, Networking, Strage and Analysis, SC12. https://science.energy.gov/~/media/ascr/ascac/pdf/reports/2013/SC12_Harrod.pdf
- 10.White paper: FUJITSU Supercomputer PRIMEHPC FX100—Evolution to the Next Generation. https://www.fujitsu.com/global/Images/primehpc-fx100-hard-en.pdf
- 11.MacKerell AD, Bashford D, Bellott M, Dunbrack RL, Evanseck JD, Field MF, Fischer S, Gao J, Guo H, Ha S, Joseph-McCarthy D, Kuchnir L, Kuczera K, Lau FT, Mattos C, Michnick S, Ngo T, Nguyen DT, Prodhom B, Reiher WE, Roux B, Schlenkrich M, Smith JC, Stote R, Straub J, Watanabe M, Wiorkiewicz-Kuczera J, Yin D, Karplus M (1998) All-atom empirical potential for molecular modeling and dynamics studies of proteins. J Phys Chem B 102:3586–3616CrossRefGoogle Scholar
- 25.Cundall PA (1971) A computer model for simulating progressive large scale movements in blocky rock Systems. In: Proceedings of the Symposium of the International Society of Rock Mechanics, Nancy, FranceGoogle Scholar