Parallel Solvers on the GPU for Large-Scale Finite Element Equations

We present a parallel solution based on CUDA for accelerating the computation for solving large-scale Finite Element equations in electrical and magnetic field. JCG is used for solving equations and corresponding kernel function is designed for spMV. A computation speed test for solving FE equations...

Full description

Saved in:
Bibliographic Details
Published inApplied Mechanics and Materials Vol. 701-702; no. Industrial Engineering, Computation and Information Technologies; pp. 207 - 213
Main Authors Wang, Dong, Chen, Jun Quan, Zhang, Qing Hu, Jiang, Ya Peng
Format Journal Article
LanguageEnglish
Published Zurich Trans Tech Publications Ltd 01.12.2014
Subjects
Central processing units
Computation
Finite element method
Hardware
Magnetic fields
Mathematical analysis
Mathematical models
Solvers
CUDA
Finite Element Equations
Jacobi Conjugate Gradient Method
Parallel
GPU
Online AccessGet full text
ISBN9783038353485
3038353485
ISSN1660-9336
1662-7482
1662-7482
DOI10.4028/www.scientific.net/AMM.701-702.207

Cover

Abstract We present a parallel solution based on CUDA for accelerating the computation for solving large-scale Finite Element equations in electrical and magnetic field. JCG is used for solving equations and corresponding kernel function is designed for spMV. A computation speed test for solving FE equations is taken on NVIDIA Tesla K20c GPU hardware platform, the result proves that the method of kernel can reach 17.1 times faster than the solution using CPU, however it cannot ensure the advantage with CPU if we only use the lib functions on GPU to solve equations.
AbstractList We present a parallel solution based on CUDA for accelerating the computation for solving large-scale Finite Element equations in electrical and magnetic field. JCG is used for solving equations and corresponding kernel function is designed for spMV. A computation speed test for solving FE equations is taken on NVIDIA Tesla K20c GPU hardware platform, the result proves that the method of kernel can reach 17.1 times faster than the solution using CPU, however it cannot ensure the advantage with CPU if we only use the lib functions on GPU to solve equations.
Author Jiang, Ya Peng
Wang, Dong
Zhang, Qing Hu
Chen, Jun Quan
Author_xml – givenname: Dong
  surname: Wang
  fullname: Wang, Dong
  email: wangdongl@vip.sina.com
  organization: Naval University of Engineering : National Key Laboratory for Vessel Integrated Power System Technology
– givenname: Jun Quan
  surname: Chen
  fullname: Chen, Jun Quan
  email: 303973172@qq.com
  organization: Naval University of Engineering : National Key Laboratory for Vessel Integrated Power System Technology
– givenname: Qing Hu
  surname: Zhang
  fullname: Zhang, Qing Hu
  email: roywade90@126.com
  organization: Naval University of Engineering : National Key Laboratory for Vessel Integrated Power System Technology
– givenname: Ya Peng
  surname: Jiang
  fullname: Jiang, Ya Peng
  email: 535749504@qq.com
  organization: Naval University of Engineering : National Key Laboratory for Vessel Integrated Power System Technology
BookMark eNqNkE1LJDEQhsPqgjr6HwJeROi2knR3kqMfo7swsoJ6DulM9dqSSTTpcfDfG3cWdtnTHoo61MNbL88B2Q0xICGnDOoGuDrbbDZ1diOGaRxGVweczs5vb2sJrJLAaw7yC9lnXccr2Si-Q460VAKEEq1oVLv76waVFqLbIwc5PwN0DWvUPrm6s8l6j57eR_-GKdMY6PSE9ObukQ4x0YVNP7G6d9YjvR7DOCGde1yVJnT-urbTGEM-JF8H6zMe_d4z8ng9f7j8Vi1-3Hy_PF9Ujikmq2bJsYwGpwcE5BJsoy32FrRsmO4BuOyXfbvkbd_3cpBMDUIPpWvnWmFRzMjJNvclxdc15smsxuzQexswrrNhXcuEllJCQY__QZ_jOoXSrlCCcdXKYmhGLraUSzHnhIN5SePKpnfDwHyKN0W8-SPeFPGmiDdFfBluivgScrUNmZINeUL39Nev_4_5AGTwk7Y
Cites_doi 10.1109/tmag.2010.2043511
10.3724/sp.j.1087.2013.00825
10.1109/tmag.2011.2173304
ContentType Journal Article
Copyright 2015 Trans Tech Publications Ltd
Copyright Trans Tech Publications Ltd. Dec 2014
Copyright_xml – notice: 2015 Trans Tech Publications Ltd
– notice: Copyright Trans Tech Publications Ltd. Dec 2014
DBID AAYXX
CITATION
7SR
7TB
8BQ
8FD
8FE
8FG
ABJCF
ABUWG
AFKRA
BENPR
BFMQW
BGLVJ
CCPQU
D1I
DWQXO
FR3
HCIFZ
JG9
KB.
KR7
L6V
M7S
PDBOC
PHGZM
PHGZT
PKEHL
PQEST
PQGLB
PQQKQ
PQUKI
PRINS
PTHSS
DOI 10.4028/www.scientific.net/AMM.701-702.207
DatabaseName CrossRef
Engineered Materials Abstracts
Mechanical & Transportation Engineering Abstracts
METADEX
Technology Research Database
ProQuest SciTech Collection
ProQuest Technology Collection
Materials Science & Engineering Collection
ProQuest Central (Alumni)
ProQuest Central UK/Ireland
ProQuest Central
Continental Europe Database
Technology collection
ProQuest One Community College
ProQuest Materials Science Collection
ProQuest Central
Engineering Research Database
SciTech Premium Collection
Materials Research Database
Materials Science Database
Civil Engineering Abstracts
ProQuest Engineering Collection
Engineering Database
Materials Science Collection
ProQuest Central Premium
ProQuest One Academic
ProQuest One Academic Middle East (New)
ProQuest One Academic Eastern Edition (DO NOT USE)
ProQuest One Applied & Life Sciences
ProQuest One Academic
ProQuest One Academic UKI Edition
ProQuest Central China
Engineering Collection
DatabaseTitle CrossRef
Materials Research Database
Technology Collection
Technology Research Database
ProQuest One Academic Middle East (New)
Mechanical & Transportation Engineering Abstracts
Materials Science Collection
ProQuest Central (Alumni Edition)
SciTech Premium Collection
ProQuest One Community College
ProQuest Central China
ProQuest Central
ProQuest One Applied & Life Sciences
Engineered Materials Abstracts
ProQuest Engineering Collection
ProQuest Central Korea
Materials Science Database
ProQuest Central (New)
Engineering Collection
ProQuest Materials Science Collection
Civil Engineering Abstracts
Engineering Database
ProQuest One Academic Eastern Edition
ProQuest Technology Collection
Continental Europe Database
ProQuest SciTech Collection
METADEX
ProQuest One Academic UKI Edition
Materials Science & Engineering Collection
Engineering Research Database
ProQuest One Academic
ProQuest One Academic (New)
DatabaseTitleList
Civil Engineering Abstracts
Materials Research Database
CrossRef
Database_xml – sequence: 1
  dbid: 8FG
  name: ProQuest Technology Collection
  url: https://search.proquest.com/technologycollection1
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 1662-7482
EndPage 213
ExternalDocumentID 3517461661
10_4028_www_scientific_net_AMM_701_702_207
GroupedDBID .DC
4.4
6J9
8FE
8FG
ABHXD
ABJCF
ABJNI
ABUWG
ACGFO
ACGFS
ACIWK
AFKRA
ALMA_UNASSIGNED_HOLDINGS
BENPR
BFMQW
BGLVJ
BPHCQ
CCPQU
CZ9
D1I
DB1
DKFMR
EBS
EJD
HCIFZ
KB.
KC.
L6V
M7S
P2P
PDBOC
PHGZM
PHGZT
PQGLB
PQQKQ
PROAC
PTHSS
RNS
RTP
AAYXX
ABDNZ
ACYGS
CITATION
PUEGO
7SR
7TB
8BQ
8FD
DWQXO
FR3
JG9
KR7
PKEHL
PQEST
PQUKI
PRINS
ID FETCH-LOGICAL-c1817-4d2e4d290c9fe0e270a49aeba097419b0027bdb5d25bbb7f718f39f0066c53ae3
IEDL.DBID BENPR
ISBN 9783038353485
3038353485
ISSN 1660-9336
1662-7482
IngestDate Thu Oct 02 11:53:34 EDT 2025
Fri Jul 25 11:50:37 EDT 2025
Wed Oct 01 04:00:46 EDT 2025
Fri Oct 03 21:33:11 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue Industrial Engineering, Computation and Information Technologies
Keywords CUDA
Finite Element Equations
Jacobi Conjugate Gradient Method
Parallel
GPU
Language English
License https://www.scientific.net/PolicyAndEthics/PublishingPolicies
https://www.scientific.net/license/TDM_Licenser.pdf
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c1817-4d2e4d290c9fe0e270a49aeba097419b0027bdb5d25bbb7f718f39f0066c53ae3
Notes Selected, peer reviewed papers from the 2014 2nd International Conference on Mechatronics and Information Technology (ICMIT 2014), October 18-19, 2014, Chongqing, China
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
PQID 1631285778
PQPubID 2029177
PageCount 7
ParticipantIDs proquest_miscellaneous_1651397770
proquest_journals_1631285778
crossref_primary_10_4028_www_scientific_net_AMM_701_702_207
transtech_journals_10_4028_www_scientific_net_AMM_701_702_207
ProviderPackageCode CITATION
AAYXX
PublicationCentury 2000
PublicationDate 20141201
PublicationDateYYYYMMDD 2014-12-01
PublicationDate_xml – month: 12
  year: 2014
  text: 20141201
  day: 01
PublicationDecade 2010
PublicationPlace Zurich
PublicationPlace_xml – name: Zurich
PublicationTitle Applied Mechanics and Materials
PublicationYear 2014
Publisher Trans Tech Publications Ltd
Publisher_xml – name: Trans Tech Publications Ltd
References 499791
499790
499793
499792
499784
499795
499794
499786
499785
499788
499787
499789
References_xml – ident: 499785
  doi: 10.1109/tmag.2010.2043511
– ident: 499792
  doi: 10.3724/sp.j.1087.2013.00825
– ident: 499795
– ident: 499784
– ident: 499794
– ident: 499787
– ident: 499788
– ident: 499789
– ident: 499791
– ident: 499786
  doi: 10.1109/tmag.2011.2173304
– ident: 499790
– ident: 499793
SSID ssj0064148
ssj0001433968
Score 1.9533315
Snippet We present a parallel solution based on CUDA for accelerating the computation for solving large-scale Finite Element equations in electrical and magnetic...
SourceID proquest
crossref
transtech
SourceType Aggregation Database
Index Database
Publisher
StartPage 207
SubjectTerms Central processing units
Computation
Finite element method
Hardware
Magnetic fields
Mathematical analysis
Mathematical models
Solvers
Title Parallel Solvers on the GPU for Large-Scale Finite Element Equations
URI https://www.scientific.net/AMM.701-702.207
https://www.proquest.com/docview/1631285778
https://www.proquest.com/docview/1651397770
Volume 701-702
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVPQU
  databaseName: Continental Europe Database
  isbn: 9783038353485
  customDbUrl:
  eissn: 1662-7482
  dateEnd: 20200630
  omitProxy: false
  ssIdentifier: ssj0064148
  issn: 1660-9336
  databaseCode: BFMQW
  dateStart: 20040901
  isFulltext: true
  titleUrlDefault: https://search.proquest.com/conteurope
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: ProQuest Central
  isbn: 9783038353485
  customDbUrl: http://www.proquest.com/pqcentral?accountid=15518
  eissn: 1662-7482
  dateEnd: 20200630
  omitProxy: true
  ssIdentifier: ssj0064148
  issn: 1660-9336
  databaseCode: BENPR
  dateStart: 20040901
  isFulltext: true
  titleUrlDefault: https://www.proquest.com/central
  providerName: ProQuest
– providerCode: PRVPQU
  databaseName: ProQuest Technology Collection
  isbn: 9783038353485
  customDbUrl:
  eissn: 1662-7482
  dateEnd: 20241105
  omitProxy: true
  ssIdentifier: ssj0064148
  issn: 1660-9336
  databaseCode: 8FG
  dateStart: 20040901
  isFulltext: true
  titleUrlDefault: https://search.proquest.com/technologycollection1
  providerName: ProQuest
– providerCode: PRVJCC
  databaseName: Ebook Central - Academic Complete
  isbn: 9783038353485
  customDbUrl:
  eissn: 1662-7482
  dateEnd: 99999999
  omitProxy: false
  ssIdentifier: ssj0001433968
  issn: 1660-9336
  databaseCode: ADVEM
  dateStart: 0
  isFulltext: true
  titleUrlDefault: https://www.proquest.com/libraries/academic/ebooks/
  providerName: ProQuest Ebooks
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV3dS9xAEB_0Dlp9EG0rnh_HCj4JaZNskk0QET8SpTTHtfaob0s22YAgiXrn_-9MNjHXF_EhhJAPll82M7_ZzPwG4IiLKC_ygMISamFWBpGl0JFZKoxIXQ2tg0PFyekkuJl5P-_8uxWYdLUwlFbZ2cTGUBd1TmvkP5A3oCn1hQjPHp8s6hpFf1e7FhpZ21qhOG0kxlZh6JIy1gCGF_Fk-qdfdfE4b9RgjK0OPKfpr-UEOFoM7QOq_rAxaPO5F_pGn6c__gTHaGMw4AqbYZqKRUroaVYRztP0u8DY3NR3if_9XE9ePy_IB5FA65IfSzZhoyWg7NzMmC1Y0dUXWF-SJfwKV9PsmVqsPLDbmhKn56yuGBJFdj2dMSS57Belj1u3-Ho1S-6JtbLYpKGz-Mloh8-_wSyJ_17eWG23BStHLy8sr3A1bpGdR6W2tSvszIsyrTIbQw6HpBNdoQrlF66vlBIlOrWSRyVxltznmebbMKjqSu8AC_ECkn3hRNB4jhzHofsyCs_csCxGcNJhIx-NqIbEYISQlYis7JGViKxEZCUii5srEdkR7HdwyvaDm8t-eozg8O00fir0_yOrdP1C1_gN3xX2CE7fXsPSMz48hN33h7AHa8irPJP1sg-DxfOLPkDuslBjWA2T6zEMkXTE6bidnrRP0t__XgED-erY
linkProvider ProQuest
linkToHtml http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3fa9RAEB5qC60-SP2Fp1VXsC-FaJLdZLPIIbW982rvjsP2oG9rNtmAIEnbuyL-c_3bOpNNmuuL-NKHPIT8IHzZnflmd-YbgA9cqizPYgpLqIVZESvPoCPzTKJIXQ2tQ0DFyZNpPJqL72fR2Rpct7UwlFbZ2sTaUOdVRmvkn5A3oCmNpEy-nF941DWKdlfbFhpp01oh79cSY01hx7H9-wdDuEX_6BD_924YDgenByOv6TLgZejdpCfy0OKh_EwV1reh9FOhUmtSH6l2QJKBoTS5ifIwMsbIAo15wVVBvjqLeGo5vvcBbAguFAZ_G18H09mPbpVHcF6rzzjfEIug7ucVxIiO4jymahMfg8SIiyRyekDd-SbsoU3DAC-pYXEVkpRAVK9a7E8mH6UfeK6eTN71qx1Z3lqSzyNB2BW_OdyGxw3hZftuhD6BNVs-hUcrMojP4HCWXlJLl9_spKJE7QWrSobElH2bzRmSajamdHXvBIeTZcNfxJLZwKW9s8GF0ypfPIf5veD-AtbLqrQvgSV4A8nMcCKEPENOFdBzKYWDYVLkPfjcYqPPnYiHxuCHkNWIrO6Q1YisRmQ1IotHqBHZHuy0cOpmgi90Nxx78P72Mk5N2m9JS1td0T1Rza-l34P-7W9Yecd_f8Krf3_CO9ganU7Genw0PX4ND5HTCZdxswPry8sr-wZ509K8bQYng5_3PR9uAIvtI7U
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Parallel+Solvers+on+the+GPU+for+Large-Scale+Finite+Element+Equations&rft.jtitle=Applied+mechanics+and+materials&rft.au=Wang%2C+Dong&rft.au=Chen%2C+Jun+Quan&rft.au=Zhang%2C+Qing+Hu&rft.au=Jiang%2C+Ya+Peng&rft.date=2014-12-01&rft.pub=Trans+Tech+Publications+Ltd&rft.issn=1660-9336&rft.eissn=1662-7482&rft.volume=701-702&rft.spage=207&rft.epage=213&rft_id=info:doi/10.4028%2Fwww.scientific.net%2FAMM.701-702.207&rft.externalDocID=10_4028_www_scientific_net_AMM_701_702_207
thumbnail_s http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=https%3A%2F%2Fwww.scientific.net%2FImage%2FTitleCover%2F3682%3Fwidth%3D600