A Performance Model for GPU-Accelerated FDTD Applications

Baumeister, Paul F.; Hater, Thorsten; Wahl, Pierre; Pleiter, Dirk; Kraus, Jiri

doi:10.1109/HiPC.2015.24

Contribution to a conference proceedings/Contribution to a book

FZJ-2016-06471

A Performance Model for GPU-Accelerated FDTD Applications

Baumeister, P. F. (Corresponding author)FZJ* ; Hater, T.FZJ* ; Kraus, J.FZJ* ; Pleiter, D.FZJ* ; Wahl, P.

2015
IEEE

2015 IEEE 22nd International Conference on High Performance Computing (HiPC) : [Proceedings] - IEEE, 2015. - ISBN 978-1-4673-8488-9
2015 IEEE 22nd International Conference on High Performance Computing (HiPC), Bengaluru, India, 16 Dec 2015 - 19 Dec 2015 IEEE 185 - 193 (2015) [10.1109/HiPC.2015.24]

This record in other databases:

Please use a persistent id in citations: doi:10.1109/HiPC.2015.24

Abstract: In this work we develop, validate and use a performance model for a Finite-Difference Time-Domain (FDTD) application which is parallelized on multiple GPUs. FDTD is a method for simulating electrodynamic interaction and is applied in a number of research and engineering areas. In this work we focus on a particular implementation called B-CALM (Belgium-California Light Machine). We adopt a simple, semi-empirical modelling approach to design a model which we validate for different hardware architectures. Using the model allows making implementation decisions and exploring the architectural design space with the goal of optimizing HPC systems for this application.

Contributing Institute(s):