A Scalable Neighbor List Construction Method for Molecular Dynamics Simulations on Multi-core Platforms

DOI

10.2991/iiicec-15.2015.38How to use a DOI?

Keywords

Parallel Computing; Molecular Dynamics; Spatial Decomposition

Abstract

We present a scalable method to implement the construction of neighbor list in molecular dynamic simulations on multi-core platforms. Our methodology is based on the OpenMP programming model. Our method doesn’t divide the whole of simulation region into cells. Each thread builds and maintains a few sliding cells for a neighbor list constructions. It reduces memory requirements and code complexity. Compared with some other methods applied in the neighbor list construction, the results of our experiments indicate that it is a scalable and efficient method.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yali Liu
AU  - Wenyan Chai
AU  - Xiurong Li
PY  - 2015/03
DA  - 2015/03
TI  - A Scalable Neighbor List Construction Method for Molecular Dynamics Simulations on Multi-core Platforms
BT  - Proceedings of the 2015 International Industrial Informatics and Computer Engineering Conference
PB  - Atlantis Press
SP  - 158
EP  - 161
SN  - 2352-538X
UR  - https://doi.org/10.2991/iiicec-15.2015.38
DO  - 10.2991/iiicec-15.2015.38
ID  - Liu2015/03
ER  -