To read this content please select one of the options below:

Parallel fast-neighbor-searching and communication strategy for particle-based methods

Lin Fu (Department of Mechanical Engineering, Technical University of Munich, Munich, Germany)
Zhe Ji (Department of Mechanical Engineering, Technical University of Munich, Munich, Germany)
Xiangyu Y. Hu (Department of Mechanical Engineering, Technical University of Munich, Munich, Germany)
Nikolaus A. Adams (Department of Mechanical Engineering, Technical University of Munich, Munich, Germany)

Engineering Computations

ISSN: 0264-4401

Article publication date: 14 March 2019

Issue publication date: 8 May 2019

181

Abstract

Purpose

This paper aims to develop a parallel fast neighbor search method and communication strategy for particle-based methods with adaptive smoothing-length on distributed-memory computing systems.

Design/methodology/approach

With a multi-resolution-based hierarchical data structure, the parallel neighbor search method is developed to detect and construct ghost buffer particles, i.e. neighboring particles on remote processor nodes. To migrate ghost buffer particles among processor nodes, an undirected graph is established to characterize the sparse data communication relation and is dynamically recomposed. By the introduction of an edge coloring algorithm from graph theory, the complex sparse data exchange can be accomplished within optimized frequency. For each communication substep, only efficient nonblocking point-to-point communication is involved.

Findings

Two demonstration scenarios are considered: fluid dynamics based on smoothed-particle hydrodynamics with adaptive smoothing-length and a recently proposed physics-motivated partitioning method [Fu et al., JCP 341 (2017): 447-473]. Several new concepts are introduced to recast the partitioning method into a parallel version. A set of numerical experiments is conducted to demonstrate the performance and potential of the proposed parallel algorithms.

Originality/value

The proposed methods are simple to implement in large-scale parallel environment and can handle particle simulations with arbitrarily varying smoothing-lengths. The implemented smoothed-particle hydrodynamics solver has good parallel performance, suggesting the potential for other scientific applications.

Keywords

Acknowledgements

The first author is partially supported by China Scholarship Council (NO. 201206290022). The second author is partially supported by China Scholarship Council (NO. 201506290038). The authors acknowledge the useful discussions with Dr Sergey Litvinov about the edge-coloring algorithm. The computational resources are provided by Leibniz-Rechenzentrum der Bayerischen Akademie der Wissenschaften, München (LRZ).

Citation

Fu, L., Ji, Z., Hu, X.Y. and Adams, N.A. (2019), "Parallel fast-neighbor-searching and communication strategy for particle-based methods", Engineering Computations, Vol. 36 No. 3, pp. 899-929. https://doi.org/10.1108/EC-05-2018-0226

Publisher

:

Emerald Publishing Limited

Copyright © 2019, Emerald Publishing Limited

Related articles