Graphics processing unit (GPU) accelerated fast multipole BEM with level-skip M2L for 3D elasticity problems

Yingjun Wang; Qifu Wang; Xiaowei Deng; Zhaohui Xia; Jinhui Yan; Hua Xu

doi:10.1016/j.advengsoft.2015.01.002

Graphics processing unit (GPU) accelerated fast multipole BEM with level-skip M2L for 3D elasticity problems

Yingjun Wang, Qifu Wang, Xiaowei Deng, Zhaohui Xia, Jinhui Yan, Hua Xu

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

Abstract

In order to accelerate fast multipole boundary element method (FMBEM), in terms of the intrinsic parallelism of boundary elements and the FMBEM tree structure, a series of CUDA based GPU parallel algorithms for different parts of FMBEM with level-skip M2L for 3D elasticity are presented. A rigid body motion method (RBMM) for the FMBEM is proposed based on special displacement boundary conditions to deal with strongly singular integration and free term coefficients. The numerical example results show that our parallel algorithms obviously accelerates the FMBEM and can be used in large scale engineering problems with wide applications in the future.

Original language	English (US)
Pages (from-to)	105-118
Number of pages	14
Journal	Advances in Engineering Software
Volume	82
DOIs	https://doi.org/10.1016/j.advengsoft.2015.01.002
State	Published - Apr 2015
Externally published	Yes

Keywords

3D elasticity
Boundary element method
Fast multipole method
Graphics processing unit
Level-skip M2L
Parallel computing

ASJC Scopus subject areas

Software
Engineering(all)

Online availability

10.1016/j.advengsoft.2015.01.002

Library availability

Discover UIUC Full Text

Cite this

@article{6e61a6f7b497438d873bdc6a55363281,

title = "Graphics processing unit (GPU) accelerated fast multipole BEM with level-skip M2L for 3D elasticity problems",

abstract = "In order to accelerate fast multipole boundary element method (FMBEM), in terms of the intrinsic parallelism of boundary elements and the FMBEM tree structure, a series of CUDA based GPU parallel algorithms for different parts of FMBEM with level-skip M2L for 3D elasticity are presented. A rigid body motion method (RBMM) for the FMBEM is proposed based on special displacement boundary conditions to deal with strongly singular integration and free term coefficients. The numerical example results show that our parallel algorithms obviously accelerates the FMBEM and can be used in large scale engineering problems with wide applications in the future.",

keywords = "3D elasticity, Boundary element method, Fast multipole method, Graphics processing unit, Level-skip M2L, Parallel computing",

author = "Yingjun Wang and Qifu Wang and Xiaowei Deng and Zhaohui Xia and Jinhui Yan and Hua Xu",

note = "Funding Information: This research has been supported by two National Natural Science Foundations of China ( 51475180 and 51275182 ), and also sponsored in part by the National Key Technology R & D Program of China (Grants: 2012BAF08B01 ) and NSF grant CMMI-1068106 . Publisher Copyright: {\textcopyright} 2015 Elsevier Ltd. All rights reserved.",

year = "2015",

month = apr,

doi = "10.1016/j.advengsoft.2015.01.002",

language = "English (US)",

volume = "82",

pages = "105--118",

journal = "Advances in Engineering Software",

issn = "0965-9978",

publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Graphics processing unit (GPU) accelerated fast multipole BEM with level-skip M2L for 3D elasticity problems

AU - Wang, Yingjun

AU - Wang, Qifu

AU - Deng, Xiaowei

AU - Xia, Zhaohui

AU - Yan, Jinhui

AU - Xu, Hua

N1 - Funding Information: This research has been supported by two National Natural Science Foundations of China ( 51475180 and 51275182 ), and also sponsored in part by the National Key Technology R & D Program of China (Grants: 2012BAF08B01 ) and NSF grant CMMI-1068106 . Publisher Copyright: © 2015 Elsevier Ltd. All rights reserved.

PY - 2015/4

Y1 - 2015/4

N2 - In order to accelerate fast multipole boundary element method (FMBEM), in terms of the intrinsic parallelism of boundary elements and the FMBEM tree structure, a series of CUDA based GPU parallel algorithms for different parts of FMBEM with level-skip M2L for 3D elasticity are presented. A rigid body motion method (RBMM) for the FMBEM is proposed based on special displacement boundary conditions to deal with strongly singular integration and free term coefficients. The numerical example results show that our parallel algorithms obviously accelerates the FMBEM and can be used in large scale engineering problems with wide applications in the future.

AB - In order to accelerate fast multipole boundary element method (FMBEM), in terms of the intrinsic parallelism of boundary elements and the FMBEM tree structure, a series of CUDA based GPU parallel algorithms for different parts of FMBEM with level-skip M2L for 3D elasticity are presented. A rigid body motion method (RBMM) for the FMBEM is proposed based on special displacement boundary conditions to deal with strongly singular integration and free term coefficients. The numerical example results show that our parallel algorithms obviously accelerates the FMBEM and can be used in large scale engineering problems with wide applications in the future.

KW - 3D elasticity

KW - Boundary element method

KW - Fast multipole method

KW - Graphics processing unit

KW - Level-skip M2L

KW - Parallel computing

UR - http://www.scopus.com/inward/record.url?scp=84922450109&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84922450109&partnerID=8YFLogxK

U2 - 10.1016/j.advengsoft.2015.01.002

DO - 10.1016/j.advengsoft.2015.01.002

M3 - Article

AN - SCOPUS:84922450109

SN - 0965-9978

VL - 82

SP - 105

EP - 118

JO - Advances in Engineering Software

JF - Advances in Engineering Software

ER -

Graphics processing unit (GPU) accelerated fast multipole BEM with level-skip M2L for 3D elasticity problems

Abstract

Keywords

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this