Solving the Klein-Gordon equation using Fourier spectral methods

a benchmark test for computer performance

S. Aseeri, O. Batrašev, M. Icardi, B. Leu, A. Liu, N. Li, B. K. Muite, Eike Müller, B. Palen, M. Quell, H. Servat, P. Sheth, R. Speck, M. Van Moer, J. Vienne

Research output: Chapter in Book/Report/Conference proceedingChapter

2 Citations (Scopus)
34 Downloads (Pure)

Abstract

The cubic Klein-Gordon equation is a simple but non-trivial partial differential equation whose numerical solution has the main building blocks required for the solution of many other partial differential equations. In this study, the library 2DECOMP&FFT is used in a Fourier spectral scheme to solve the Klein-Gordon equation and strong scaling of the code is examined on thirteen different machines for a problem size of 512^3. The results are useful in assessing likely performance of other parallel fast Fourier transform based programs for solving partial differential equations. The problem is chosen to be large enough to solve on a workstation, yet also of interest to solve quickly on a supercomputer, in particular for parametric studies. Unlike the Linpack benchmark, a high ranking will not be obtained by simply building a bigger computer.
Original languageEnglish
Title of host publicationHPC '15 Proceedings of the Symposium on High Performance Computing
Place of PublicationSan Diego, U. S. A.
PublisherSociety for Computer Simulation International
Pages182-191
Number of pages10
ISBN (Print)9781510801011
Publication statusPublished - 12 Apr 2015

Fingerprint

Fourier Method
Klein-Gordon Equation
Spectral Methods
Benchmark
Partial differential equation
Cubic equation
Supercomputer
Fast Fourier transform
Building Blocks
Ranking
Likely
Numerical Solution
Scaling
Differential equation

Keywords

  • cs.PF
  • cs.DC
  • math.NA

Cite this

Aseeri, S., Batrašev, O., Icardi, M., Leu, B., Liu, A., Li, N., ... Vienne, J. (2015). Solving the Klein-Gordon equation using Fourier spectral methods: a benchmark test for computer performance. In HPC '15 Proceedings of the Symposium on High Performance Computing (pp. 182-191). San Diego, U. S. A.: Society for Computer Simulation International .

Solving the Klein-Gordon equation using Fourier spectral methods : a benchmark test for computer performance. / Aseeri, S.; Batrašev, O.; Icardi, M.; Leu, B.; Liu, A.; Li, N.; Muite, B. K.; Müller, Eike; Palen, B.; Quell, M.; Servat, H.; Sheth, P.; Speck, R.; Moer, M. Van; Vienne, J.

HPC '15 Proceedings of the Symposium on High Performance Computing . San Diego, U. S. A. : Society for Computer Simulation International , 2015. p. 182-191.

Research output: Chapter in Book/Report/Conference proceedingChapter

Aseeri, S, Batrašev, O, Icardi, M, Leu, B, Liu, A, Li, N, Muite, BK, Müller, E, Palen, B, Quell, M, Servat, H, Sheth, P, Speck, R, Moer, MV & Vienne, J 2015, Solving the Klein-Gordon equation using Fourier spectral methods: a benchmark test for computer performance. in HPC '15 Proceedings of the Symposium on High Performance Computing . Society for Computer Simulation International , San Diego, U. S. A., pp. 182-191.
Aseeri S, Batrašev O, Icardi M, Leu B, Liu A, Li N et al. Solving the Klein-Gordon equation using Fourier spectral methods: a benchmark test for computer performance. In HPC '15 Proceedings of the Symposium on High Performance Computing . San Diego, U. S. A.: Society for Computer Simulation International . 2015. p. 182-191
Aseeri, S. ; Batrašev, O. ; Icardi, M. ; Leu, B. ; Liu, A. ; Li, N. ; Muite, B. K. ; Müller, Eike ; Palen, B. ; Quell, M. ; Servat, H. ; Sheth, P. ; Speck, R. ; Moer, M. Van ; Vienne, J. / Solving the Klein-Gordon equation using Fourier spectral methods : a benchmark test for computer performance. HPC '15 Proceedings of the Symposium on High Performance Computing . San Diego, U. S. A. : Society for Computer Simulation International , 2015. pp. 182-191
@inbook{97231cd3a01f49ca8a2f2c94995225d6,
title = "Solving the Klein-Gordon equation using Fourier spectral methods: a benchmark test for computer performance",
abstract = "The cubic Klein-Gordon equation is a simple but non-trivial partial differential equation whose numerical solution has the main building blocks required for the solution of many other partial differential equations. In this study, the library 2DECOMP&FFT is used in a Fourier spectral scheme to solve the Klein-Gordon equation and strong scaling of the code is examined on thirteen different machines for a problem size of 512^3. The results are useful in assessing likely performance of other parallel fast Fourier transform based programs for solving partial differential equations. The problem is chosen to be large enough to solve on a workstation, yet also of interest to solve quickly on a supercomputer, in particular for parametric studies. Unlike the Linpack benchmark, a high ranking will not be obtained by simply building a bigger computer.",
keywords = "cs.PF, cs.DC, math.NA",
author = "S. Aseeri and O. Batrašev and M. Icardi and B. Leu and A. Liu and N. Li and Muite, {B. K.} and Eike M{\"u}ller and B. Palen and M. Quell and H. Servat and P. Sheth and R. Speck and Moer, {M. Van} and J. Vienne",
year = "2015",
month = "4",
day = "12",
language = "English",
isbn = "9781510801011",
pages = "182--191",
booktitle = "HPC '15 Proceedings of the Symposium on High Performance Computing",
publisher = "Society for Computer Simulation International",

}

TY - CHAP

T1 - Solving the Klein-Gordon equation using Fourier spectral methods

T2 - a benchmark test for computer performance

AU - Aseeri, S.

AU - Batrašev, O.

AU - Icardi, M.

AU - Leu, B.

AU - Liu, A.

AU - Li, N.

AU - Muite, B. K.

AU - Müller, Eike

AU - Palen, B.

AU - Quell, M.

AU - Servat, H.

AU - Sheth, P.

AU - Speck, R.

AU - Moer, M. Van

AU - Vienne, J.

PY - 2015/4/12

Y1 - 2015/4/12

N2 - The cubic Klein-Gordon equation is a simple but non-trivial partial differential equation whose numerical solution has the main building blocks required for the solution of many other partial differential equations. In this study, the library 2DECOMP&FFT is used in a Fourier spectral scheme to solve the Klein-Gordon equation and strong scaling of the code is examined on thirteen different machines for a problem size of 512^3. The results are useful in assessing likely performance of other parallel fast Fourier transform based programs for solving partial differential equations. The problem is chosen to be large enough to solve on a workstation, yet also of interest to solve quickly on a supercomputer, in particular for parametric studies. Unlike the Linpack benchmark, a high ranking will not be obtained by simply building a bigger computer.

AB - The cubic Klein-Gordon equation is a simple but non-trivial partial differential equation whose numerical solution has the main building blocks required for the solution of many other partial differential equations. In this study, the library 2DECOMP&FFT is used in a Fourier spectral scheme to solve the Klein-Gordon equation and strong scaling of the code is examined on thirteen different machines for a problem size of 512^3. The results are useful in assessing likely performance of other parallel fast Fourier transform based programs for solving partial differential equations. The problem is chosen to be large enough to solve on a workstation, yet also of interest to solve quickly on a supercomputer, in particular for parametric studies. Unlike the Linpack benchmark, a high ranking will not be obtained by simply building a bigger computer.

KW - cs.PF

KW - cs.DC

KW - math.NA

M3 - Chapter

SN - 9781510801011

SP - 182

EP - 191

BT - HPC '15 Proceedings of the Symposium on High Performance Computing

PB - Society for Computer Simulation International

CY - San Diego, U. S. A.

ER -