Paper Info
Title
SWIFT: Maintaining weak-scalability with a dynamic range of 104 in time-step size to harness extreme adaptivity.
Abstract
Cosmological simulations require the use of a multiple time-stepping scheme. Without such a scheme, cosmological simulations would be impossible due to their high level of dynamic range; over eleven orders of magnitude in density. Such a large dynamic range leads to a range of over four orders of magnitude in time-step, which presents a significant load-balancing challenge. In this work, the extreme adaptivity that cosmological simulations present is tackled in three main ways through the use of the code SWIFT. First, an adaptive mesh is used to ensure that only the relevant particles are interacted in a given time-step. Second, task-based parallelism is used to ensure efficient load-balancing within a single node, using pthreads and SIMD vectorisation. Finally, a domain decomposition strategy is presented, using the graph domain decomposition library METIS, that bisects the work that must be performed by the simulation between nodes using MPI. These three strategies are shown to give SWIFT near-perfect weak-scaling characteristics, only losing 25% performance when scaling from 1 to 4096 cores on a representative problem, whilst being more than 30x faster than the de-facto standard Gadget-2 code.
Year
Venue
Field
2018
arXiv: Distributed, Parallel, and Cluster Computing
Orders of magnitude (numbers),Dynamic range,Swift,Computer science,Parallel computing,SIMD,POSIX Threads,Scaling,Domain decomposition methods,Scalability,Distributed computing
DocType
Volume
ISSN
Journal
abs/1807.01341
Proceedings of the 13th SPHERIC International Workshop, Galway, Ireland, June 26-28 2018, pp. 44-51
Citations 
PageRank 
References 
0
0.34
6
Authors
5
Name
Order
Citations
PageRank
Josh Borrow100.34
Richard G. Bower201.01
p w draper351.81
Pedro Gonnet48913.43
Matthieu Schaller542.85