Title: Benchmarking multi-GPU communication using the shallow water equations

Authors: Daniel P. Playne; Kenneth A. Hawick

Addresses: Department of Computer Science, Massey University, Auckland, New Zealand ' Department of Computer Science, University of Hull, Hull, HU6 7RX, UK

Abstract: The shallow water model equations provide a simple yet realistic benchmark problem in computational fluid dynamics (CFD) that can be implemented on a variety of computational platforms. Graphical processing units (GPUs) can be used to accelerate such applications on a single device using a data parallel decompositional scheme or with multiple devices using a domain decompositional approach. A shallow water equation simulation is implemented on a range of modern GPU architectures and multi-GPU systems. The typical performance of these systems using the two main device-device communication methods is reported on for single- and double-precision calculations.

Keywords: shallow water equations; parallel computing; CUDA; multi-GPU; graphical processing units; GPUs; computational fluid dynamics; CFD; simulation.

DOI: 10.1504/IJBDI.2015.070596

International Journal of Big Data Intelligence, 2015 Vol.2 No.3, pp.157 - 167

Available online: 11 Jul 2015 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article