A Hybrid MPI+Threads Approach to Particle Group Finding Using Union-Find
Abstract
The Friends-of-Friends (FoF) algorithm is a standard technique used in cosmological -body simulations to identify structures. Its goal is to find clusters of particles (called groups) that are separated by at most a cut-off radius. -body simulations typically use most of the memory present on a node, leaving very little free for a FoF algorithm to run on-the-fly. We propose a new method that utilises the common Union-Find data structure and a hybrid MPI+threads approach. The algorithm can also be expressed elegantly in a task-based formalism if such a framework is used in the rest of the application. We have implemented our algorithm in the open-source cosmological code, SWIFT. Our implementation displays excellent strong- and weak-scaling behaviour on realistic problems and compares favourably (speed-up of 18x) over other methods commonly used in the -body community.
Cite
@article{arxiv.2003.11468,
title = {A Hybrid MPI+Threads Approach to Particle Group Finding Using Union-Find},
author = {James S. Willis and Matthieu Schaller and Pedro Gonnet and John C. Helly},
journal= {arXiv preprint arXiv:2003.11468},
year = {2020}
}
Comments
12 pages, 4 figures. Proceedings of the ParCo 2019 conference, Prague, Czech Republic, September 10-13th, 2019