Fast multipole methods on a cluster of GPUs for the meshless simulation of turbulence

R. Yokota, T. Narumi, R. Sakamaki, S. Kameoka, Shinnosuke Obi, Kenji Yasuoka

Research output: Contribution to journalArticle

37 Citations (Scopus)

Abstract

Recent advances in the parallelizability of fast N-body algorithms, and the programmability of graphics processing units (GPUs) have opened a new path for particle based simulations. For the simulation of turbulence, vortex methods can now be considered as an interesting alternative to finite difference and spectral methods. The present study focuses on the efficient implementation of the fast multipole method and pseudo-particle method on a cluster of NVIDIA GeForce 8800 GT GPUs, and applies this to a vortex method calculation of homogeneous isotropic turbulence. The results of the present vortex method agree quantitatively with that of the reference calculation using a spectral method. We achieved a maximum speed of 7.48 TFlops using 64 GPUs, and the cost performance was near $9.4/GFlops. The calculation of the present vortex method on 64 GPUs took 4120 s, while the spectral method on 32 CPUs took 4910 s.

Original languageEnglish
Pages (from-to)2066-2078
Number of pages13
JournalComputer Physics Communications
Volume180
Issue number11
DOIs
Publication statusPublished - 2009 Nov

Fingerprint

multipoles
spectral methods
Vortex flow
Turbulence
turbulence
vortices
simulation
homogeneous turbulence
isotropic turbulence
Program processors
costs
Graphics processing unit
Costs

Keywords

  • Fast multipole method
  • Graphics processing unit
  • Particle method
  • Pseudo-particle method

ASJC Scopus subject areas

  • Hardware and Architecture
  • Physics and Astronomy(all)

Cite this

Fast multipole methods on a cluster of GPUs for the meshless simulation of turbulence. / Yokota, R.; Narumi, T.; Sakamaki, R.; Kameoka, S.; Obi, Shinnosuke; Yasuoka, Kenji.

In: Computer Physics Communications, Vol. 180, No. 11, 11.2009, p. 2066-2078.

Research output: Contribution to journalArticle

Yokota, R. ; Narumi, T. ; Sakamaki, R. ; Kameoka, S. ; Obi, Shinnosuke ; Yasuoka, Kenji. / Fast multipole methods on a cluster of GPUs for the meshless simulation of turbulence. In: Computer Physics Communications. 2009 ; Vol. 180, No. 11. pp. 2066-2078.
@article{36ae6780f1794d00a6b21def2b17aa2d,
title = "Fast multipole methods on a cluster of GPUs for the meshless simulation of turbulence",
abstract = "Recent advances in the parallelizability of fast N-body algorithms, and the programmability of graphics processing units (GPUs) have opened a new path for particle based simulations. For the simulation of turbulence, vortex methods can now be considered as an interesting alternative to finite difference and spectral methods. The present study focuses on the efficient implementation of the fast multipole method and pseudo-particle method on a cluster of NVIDIA GeForce 8800 GT GPUs, and applies this to a vortex method calculation of homogeneous isotropic turbulence. The results of the present vortex method agree quantitatively with that of the reference calculation using a spectral method. We achieved a maximum speed of 7.48 TFlops using 64 GPUs, and the cost performance was near $9.4/GFlops. The calculation of the present vortex method on 64 GPUs took 4120 s, while the spectral method on 32 CPUs took 4910 s.",
keywords = "Fast multipole method, Graphics processing unit, Particle method, Pseudo-particle method",
author = "R. Yokota and T. Narumi and R. Sakamaki and S. Kameoka and Shinnosuke Obi and Kenji Yasuoka",
year = "2009",
month = "11",
doi = "10.1016/j.cpc.2009.06.009",
language = "English",
volume = "180",
pages = "2066--2078",
journal = "Computer Physics Communications",
issn = "0010-4655",
publisher = "Elsevier",
number = "11",

}

TY - JOUR

T1 - Fast multipole methods on a cluster of GPUs for the meshless simulation of turbulence

AU - Yokota, R.

AU - Narumi, T.

AU - Sakamaki, R.

AU - Kameoka, S.

AU - Obi, Shinnosuke

AU - Yasuoka, Kenji

PY - 2009/11

Y1 - 2009/11

N2 - Recent advances in the parallelizability of fast N-body algorithms, and the programmability of graphics processing units (GPUs) have opened a new path for particle based simulations. For the simulation of turbulence, vortex methods can now be considered as an interesting alternative to finite difference and spectral methods. The present study focuses on the efficient implementation of the fast multipole method and pseudo-particle method on a cluster of NVIDIA GeForce 8800 GT GPUs, and applies this to a vortex method calculation of homogeneous isotropic turbulence. The results of the present vortex method agree quantitatively with that of the reference calculation using a spectral method. We achieved a maximum speed of 7.48 TFlops using 64 GPUs, and the cost performance was near $9.4/GFlops. The calculation of the present vortex method on 64 GPUs took 4120 s, while the spectral method on 32 CPUs took 4910 s.

AB - Recent advances in the parallelizability of fast N-body algorithms, and the programmability of graphics processing units (GPUs) have opened a new path for particle based simulations. For the simulation of turbulence, vortex methods can now be considered as an interesting alternative to finite difference and spectral methods. The present study focuses on the efficient implementation of the fast multipole method and pseudo-particle method on a cluster of NVIDIA GeForce 8800 GT GPUs, and applies this to a vortex method calculation of homogeneous isotropic turbulence. The results of the present vortex method agree quantitatively with that of the reference calculation using a spectral method. We achieved a maximum speed of 7.48 TFlops using 64 GPUs, and the cost performance was near $9.4/GFlops. The calculation of the present vortex method on 64 GPUs took 4120 s, while the spectral method on 32 CPUs took 4910 s.

KW - Fast multipole method

KW - Graphics processing unit

KW - Particle method

KW - Pseudo-particle method

UR - http://www.scopus.com/inward/record.url?scp=70149090358&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70149090358&partnerID=8YFLogxK

U2 - 10.1016/j.cpc.2009.06.009

DO - 10.1016/j.cpc.2009.06.009

M3 - Article

VL - 180

SP - 2066

EP - 2078

JO - Computer Physics Communications

JF - Computer Physics Communications

SN - 0010-4655

IS - 11

ER -