A detailed implementation of multithreading and out-of-core computation to the conventional boundary element algorithm with minimum code changes
Autor(a) principal: | |
---|---|
Data de Publicação: | 2023 |
Outros Autores: | |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1007/s40430-023-04034-y http://hdl.handle.net/11449/249606 |
Resumo: | The boundary element method (BEM) poses some significant advantages over other numerical methods. Nevertheless, the generation of fully populated unsymmetrical matrices can induce some important problems for the conventional BEM algorithm concerning memory limitations. Furthermore, special integration procedures are used in BEM demand substantial computational power, and these characteristics can compromise BEM applicability for large problems unless more sophisticated approaches are used. A relatively straightforward way to extend BEM usability is to employ multithreading and out-of-core capabilities. This work presents detailed algorithms that can be easily implemented in conventional BEM codes. The improvements obtained are evaluated and commented. |
id |
UNSP_b7679f77f6f6b87aadd081d2fde3fb13 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/249606 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
A detailed implementation of multithreading and out-of-core computation to the conventional boundary element algorithm with minimum code changesBoundary element methodMultithreadingOut-of-coreParallelizationThe boundary element method (BEM) poses some significant advantages over other numerical methods. Nevertheless, the generation of fully populated unsymmetrical matrices can induce some important problems for the conventional BEM algorithm concerning memory limitations. Furthermore, special integration procedures are used in BEM demand substantial computational power, and these characteristics can compromise BEM applicability for large problems unless more sophisticated approaches are used. A relatively straightforward way to extend BEM usability is to employ multithreading and out-of-core capabilities. This work presents detailed algorithms that can be easily implemented in conventional BEM codes. The improvements obtained are evaluated and commented.Department of Maintenance Petrobras S.A, Mato Grosso Do SulDepartment of Mechanical Engineering Universidade Estadual Paulista (UNESP), Brasil Av., 56, São PauloDepartment of Mechanical Engineering Universidade Estadual Paulista (UNESP), Brasil Av., 56, São PauloPetrobras S.AUniversidade Estadual Paulista (UNESP)de Souza Schiara, LeandroPaschoalini, Amarildo Tabone [UNESP]2023-07-29T16:04:19Z2023-07-29T16:04:19Z2023-02-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articlehttp://dx.doi.org/10.1007/s40430-023-04034-yJournal of the Brazilian Society of Mechanical Sciences and Engineering, v. 45, n. 2, 2023.1806-36911678-5878http://hdl.handle.net/11449/24960610.1007/s40430-023-04034-y2-s2.0-85146915884Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengJournal of the Brazilian Society of Mechanical Sciences and Engineeringinfo:eu-repo/semantics/openAccess2024-07-04T20:06:15Zoai:repositorio.unesp.br:11449/249606Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T19:57:16.623291Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
A detailed implementation of multithreading and out-of-core computation to the conventional boundary element algorithm with minimum code changes |
title |
A detailed implementation of multithreading and out-of-core computation to the conventional boundary element algorithm with minimum code changes |
spellingShingle |
A detailed implementation of multithreading and out-of-core computation to the conventional boundary element algorithm with minimum code changes de Souza Schiara, Leandro Boundary element method Multithreading Out-of-core Parallelization |
title_short |
A detailed implementation of multithreading and out-of-core computation to the conventional boundary element algorithm with minimum code changes |
title_full |
A detailed implementation of multithreading and out-of-core computation to the conventional boundary element algorithm with minimum code changes |
title_fullStr |
A detailed implementation of multithreading and out-of-core computation to the conventional boundary element algorithm with minimum code changes |
title_full_unstemmed |
A detailed implementation of multithreading and out-of-core computation to the conventional boundary element algorithm with minimum code changes |
title_sort |
A detailed implementation of multithreading and out-of-core computation to the conventional boundary element algorithm with minimum code changes |
author |
de Souza Schiara, Leandro |
author_facet |
de Souza Schiara, Leandro Paschoalini, Amarildo Tabone [UNESP] |
author_role |
author |
author2 |
Paschoalini, Amarildo Tabone [UNESP] |
author2_role |
author |
dc.contributor.none.fl_str_mv |
Petrobras S.A Universidade Estadual Paulista (UNESP) |
dc.contributor.author.fl_str_mv |
de Souza Schiara, Leandro Paschoalini, Amarildo Tabone [UNESP] |
dc.subject.por.fl_str_mv |
Boundary element method Multithreading Out-of-core Parallelization |
topic |
Boundary element method Multithreading Out-of-core Parallelization |
description |
The boundary element method (BEM) poses some significant advantages over other numerical methods. Nevertheless, the generation of fully populated unsymmetrical matrices can induce some important problems for the conventional BEM algorithm concerning memory limitations. Furthermore, special integration procedures are used in BEM demand substantial computational power, and these characteristics can compromise BEM applicability for large problems unless more sophisticated approaches are used. A relatively straightforward way to extend BEM usability is to employ multithreading and out-of-core capabilities. This work presents detailed algorithms that can be easily implemented in conventional BEM codes. The improvements obtained are evaluated and commented. |
publishDate |
2023 |
dc.date.none.fl_str_mv |
2023-07-29T16:04:19Z 2023-07-29T16:04:19Z 2023-02-01 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1007/s40430-023-04034-y Journal of the Brazilian Society of Mechanical Sciences and Engineering, v. 45, n. 2, 2023. 1806-3691 1678-5878 http://hdl.handle.net/11449/249606 10.1007/s40430-023-04034-y 2-s2.0-85146915884 |
url |
http://dx.doi.org/10.1007/s40430-023-04034-y http://hdl.handle.net/11449/249606 |
identifier_str_mv |
Journal of the Brazilian Society of Mechanical Sciences and Engineering, v. 45, n. 2, 2023. 1806-3691 1678-5878 10.1007/s40430-023-04034-y 2-s2.0-85146915884 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
Journal of the Brazilian Society of Mechanical Sciences and Engineering |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1808129142548332544 |