Parallel Matrix Multiplication and LU Factorization on Ethernet-Based Clusters

Autores
Tinetti, Fernando Gustavo; Denham, Mónica Malén; De Giusti, Armando Eduardo
Año de publicación
2003
Idioma
español castellano
Tipo de recurso
documento de conferencia
Estado
versión publicada
Descripción
This work presents a simple but effective approach for two representative linear algebra operations to be solved in parallel on Ethernet-based clusters: matrix multiplication and LU matrix factorization. The main objectives of this approach are: simplicity and performance optimization. The approach is completed at a lower level by including a broadcast routine based directly on top of UDP to take advantage of the Ethernet physical broadcast facility. The performance of the proposed algorithms implemented on Ethernet-based clusters is compared with the performance obtained with the ScaLAPACK library, which is taken as having highly optimized algorithms for distributed memory parallel computers in general and clusters in particular.
Instituto de Investigación en Informática
Materia
Informática
Cluster Computing
Parallel Linear Algebra Computing
Communication Performance
Parallel algorithms
High Performance Computing
Nivel de accesibilidad
acceso abierto
Condiciones de uso
http://creativecommons.org/licenses/by/4.0/
Repositorio
SEDICI (UNLP)
Institución
Universidad Nacional de La Plata
OAI Identificador
oai:sedici.unlp.edu.ar:10915/137418
Acceder
id SEDICI_9d1dce5c4185f52041c207d6b1f709e4
oai_identifier_str oai:sedici.unlp.edu.ar:10915/137418
network_acronym_str SEDICI
repository_id_str 1329
network_name_str SEDICI (UNLP)
spelling Parallel Matrix Multiplication and LU Factorization on Ethernet-Based ClustersTinetti, Fernando GustavoDenham, Mónica MalénDe Giusti, Armando EduardoInformáticaCluster ComputingParallel Linear Algebra ComputingCommunication PerformanceParallel algorithmsHigh Performance ComputingThis work presents a simple but effective approach for two representative linear algebra operations to be solved in parallel on Ethernet-based clusters: matrix multiplication and LU matrix factorization. The main objectives of this approach are: simplicity and performance optimization. The approach is completed at a lower level by including a broadcast routine based directly on top of UDP to take advantage of the Ethernet physical broadcast facility. The performance of the proposed algorithms implemented on Ethernet-based clusters is compared with the performance obtained with the ScaLAPACK library, which is taken as having highly optimized algorithms for distributed memory parallel computers in general and clusters in particular.Instituto de Investigación en Informática2003info:eu-repo/semantics/conferenceObjectinfo:eu-repo/semantics/publishedVersionObjeto de conferenciahttp://purl.org/coar/resource_type/c_5794info:ar-repo/semantics/documentoDeConferenciaapplication/pdf431-439http://sedici.unlp.edu.ar/handle/10915/137418spainfo:eu-repo/semantics/altIdentifier/issn/0302-9743info:eu-repo/semantics/altIdentifier/issn/1611-3349info:eu-repo/semantics/altIdentifier/doi/10.1007/978-3-540-39707-6_38info:eu-repo/semantics/openAccesshttp://creativecommons.org/licenses/by/4.0/Creative Commons Attribution 4.0 International (CC BY 4.0)reponame:SEDICI (UNLP)instname:Universidad Nacional de La Platainstacron:UNLP2025-09-03T11:03:54Zoai:sedici.unlp.edu.ar:10915/137418Institucionalhttp://sedici.unlp.edu.ar/Universidad públicaNo correspondehttp://sedici.unlp.edu.ar/oai/snrdalira@sedici.unlp.edu.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:13292025-09-03 11:03:54.417SEDICI (UNLP) - Universidad Nacional de La Platafalse
dc.title.none.fl_str_mv Parallel Matrix Multiplication and LU Factorization on Ethernet-Based Clusters
title Parallel Matrix Multiplication and LU Factorization on Ethernet-Based Clusters
spellingShingle Parallel Matrix Multiplication and LU Factorization on Ethernet-Based Clusters
Tinetti, Fernando Gustavo
Informática
Cluster Computing
Parallel Linear Algebra Computing
Communication Performance
Parallel algorithms
High Performance Computing
title_short Parallel Matrix Multiplication and LU Factorization on Ethernet-Based Clusters
title_full Parallel Matrix Multiplication and LU Factorization on Ethernet-Based Clusters
title_fullStr Parallel Matrix Multiplication and LU Factorization on Ethernet-Based Clusters
title_full_unstemmed Parallel Matrix Multiplication and LU Factorization on Ethernet-Based Clusters
title_sort Parallel Matrix Multiplication and LU Factorization on Ethernet-Based Clusters
dc.creator.none.fl_str_mv Tinetti, Fernando Gustavo
Denham, Mónica Malén
De Giusti, Armando Eduardo
author Tinetti, Fernando Gustavo
author_facet Tinetti, Fernando Gustavo
Denham, Mónica Malén
De Giusti, Armando Eduardo
author_role author
author2 Denham, Mónica Malén
De Giusti, Armando Eduardo
author2_role author
author
dc.subject.none.fl_str_mv Informática
Cluster Computing
Parallel Linear Algebra Computing
Communication Performance
Parallel algorithms
High Performance Computing
topic Informática
Cluster Computing
Parallel Linear Algebra Computing
Communication Performance
Parallel algorithms
High Performance Computing
dc.description.none.fl_txt_mv This work presents a simple but effective approach for two representative linear algebra operations to be solved in parallel on Ethernet-based clusters: matrix multiplication and LU matrix factorization. The main objectives of this approach are: simplicity and performance optimization. The approach is completed at a lower level by including a broadcast routine based directly on top of UDP to take advantage of the Ethernet physical broadcast facility. The performance of the proposed algorithms implemented on Ethernet-based clusters is compared with the performance obtained with the ScaLAPACK library, which is taken as having highly optimized algorithms for distributed memory parallel computers in general and clusters in particular.
Instituto de Investigación en Informática
description This work presents a simple but effective approach for two representative linear algebra operations to be solved in parallel on Ethernet-based clusters: matrix multiplication and LU matrix factorization. The main objectives of this approach are: simplicity and performance optimization. The approach is completed at a lower level by including a broadcast routine based directly on top of UDP to take advantage of the Ethernet physical broadcast facility. The performance of the proposed algorithms implemented on Ethernet-based clusters is compared with the performance obtained with the ScaLAPACK library, which is taken as having highly optimized algorithms for distributed memory parallel computers in general and clusters in particular.
publishDate 2003
dc.date.none.fl_str_mv 2003
dc.type.none.fl_str_mv info:eu-repo/semantics/conferenceObject
info:eu-repo/semantics/publishedVersion
Objeto de conferencia
http://purl.org/coar/resource_type/c_5794
info:ar-repo/semantics/documentoDeConferencia
format conferenceObject
status_str publishedVersion
dc.identifier.none.fl_str_mv http://sedici.unlp.edu.ar/handle/10915/137418
url http://sedici.unlp.edu.ar/handle/10915/137418
dc.language.none.fl_str_mv spa
language spa
dc.relation.none.fl_str_mv info:eu-repo/semantics/altIdentifier/issn/0302-9743
info:eu-repo/semantics/altIdentifier/issn/1611-3349
info:eu-repo/semantics/altIdentifier/doi/10.1007/978-3-540-39707-6_38
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
http://creativecommons.org/licenses/by/4.0/
Creative Commons Attribution 4.0 International (CC BY 4.0)
eu_rights_str_mv openAccess
rights_invalid_str_mv http://creativecommons.org/licenses/by/4.0/
Creative Commons Attribution 4.0 International (CC BY 4.0)
dc.format.none.fl_str_mv application/pdf
431-439
dc.source.none.fl_str_mv reponame:SEDICI (UNLP)
instname:Universidad Nacional de La Plata
instacron:UNLP
reponame_str SEDICI (UNLP)
collection SEDICI (UNLP)
instname_str Universidad Nacional de La Plata
instacron_str UNLP
institution UNLP
repository.name.fl_str_mv SEDICI (UNLP) - Universidad Nacional de La Plata
repository.mail.fl_str_mv alira@sedici.unlp.edu.ar
_version_ 1842260537713360896
score 13.13397

Publicaciones similares