Parallelization of finite element analysis of nonlinear magnetic fields using GPU

Takayuki Okimura; Teruyoshi Sasayama; Norio Takahashi; Soichiro Ikuno

doi:10.1109/TMAG.2013.2244062

Parallelization of finite element analysis of nonlinear magnetic fields using GPU

Takayuki Okimura, Teruyoshi Sasayama, Norio Takahashi, Soichiro Ikuno

Research output: Contribution to journal › Article › peer-review

18 Citations (Scopus)

Abstract

The acceleration of a nonlinear magnetic field analysis by parallelizing the finite element method (FEM) is examined using the graphics processing unit (GPU). It is shown that the speedup of the magnetic field analysis is realized by parallelizing the variable preconditioned conjugate gradient (VPCG) method. The Jacobi over-relaxation (JOR) method, conjugate residual (CR) method and conjugate gradient (CG) method are also applied in the variable preconditioning. The results of computations demonstrate that VPCG using the GPU significantly improve the performance. Especially, CG applied by variable preconditioned on GPU is 39 times faster than ICCG on a CPU.

Original language	English
Article number	6514787
Pages (from-to)	1557-1560
Number of pages	4
Journal	IEEE Transactions on Magnetics
Volume	49
Issue number	5
DOIs	https://doi.org/10.1109/TMAG.2013.2244062
Publication status	Published - 2013
Externally published	Yes

All Science Journal Classification (ASJC) codes

Electronic, Optical and Magnetic Materials
Electrical and Electronic Engineering

Access to Document

10.1109/TMAG.2013.2244062

Cite this

@article{46054880974e4e23a5ddfd60b6777274,

title = "Parallelization of finite element analysis of nonlinear magnetic fields using GPU",

abstract = "The acceleration of a nonlinear magnetic field analysis by parallelizing the finite element method (FEM) is examined using the graphics processing unit (GPU). It is shown that the speedup of the magnetic field analysis is realized by parallelizing the variable preconditioned conjugate gradient (VPCG) method. The Jacobi over-relaxation (JOR) method, conjugate residual (CR) method and conjugate gradient (CG) method are also applied in the variable preconditioning. The results of computations demonstrate that VPCG using the GPU significantly improve the performance. Especially, CG applied by variable preconditioned on GPU is 39 times faster than ICCG on a CPU.",

author = "Takayuki Okimura and Teruyoshi Sasayama and Norio Takahashi and Soichiro Ikuno",

year = "2013",

doi = "10.1109/TMAG.2013.2244062",

language = "English",

volume = "49",

pages = "1557--1560",

journal = "IEEE Transactions on Magnetics",

issn = "0018-9464",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "5",

}

TY - JOUR

T1 - Parallelization of finite element analysis of nonlinear magnetic fields using GPU

AU - Okimura, Takayuki

AU - Sasayama, Teruyoshi

AU - Takahashi, Norio

AU - Ikuno, Soichiro

PY - 2013

Y1 - 2013

N2 - The acceleration of a nonlinear magnetic field analysis by parallelizing the finite element method (FEM) is examined using the graphics processing unit (GPU). It is shown that the speedup of the magnetic field analysis is realized by parallelizing the variable preconditioned conjugate gradient (VPCG) method. The Jacobi over-relaxation (JOR) method, conjugate residual (CR) method and conjugate gradient (CG) method are also applied in the variable preconditioning. The results of computations demonstrate that VPCG using the GPU significantly improve the performance. Especially, CG applied by variable preconditioned on GPU is 39 times faster than ICCG on a CPU.

AB - The acceleration of a nonlinear magnetic field analysis by parallelizing the finite element method (FEM) is examined using the graphics processing unit (GPU). It is shown that the speedup of the magnetic field analysis is realized by parallelizing the variable preconditioned conjugate gradient (VPCG) method. The Jacobi over-relaxation (JOR) method, conjugate residual (CR) method and conjugate gradient (CG) method are also applied in the variable preconditioning. The results of computations demonstrate that VPCG using the GPU significantly improve the performance. Especially, CG applied by variable preconditioned on GPU is 39 times faster than ICCG on a CPU.

UR - http://www.scopus.com/inward/record.url?scp=84877871992&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84877871992&partnerID=8YFLogxK

U2 - 10.1109/TMAG.2013.2244062

DO - 10.1109/TMAG.2013.2244062

M3 - Article

AN - SCOPUS:84877871992

SN - 0018-9464

VL - 49

SP - 1557

EP - 1560

JO - IEEE Transactions on Magnetics

JF - IEEE Transactions on Magnetics

IS - 5

M1 - 6514787

ER -

Parallelization of finite element analysis of nonlinear magnetic fields using GPU

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this