Applying backfilling over a non-dedicated cluster

Autores
Hanzich, Mauricio; Giné de Solà, Francesc; Hernández Budé, Porfidio; Solsona Theás, Francesc; Luque Fadón, Emilio
Año de publicación
2005
Idioma
inglés
Tipo de recurso
documento de conferencia
Estado
versión publicada
Descripción
The resource utilization level in open laboratories of several universities has been shown to be very low. Our aim is to take advantage of those idle resources for parallel computation without disturbing the local load. In order to provide a system that lets us execute parallel applications in such a non-dedicated cluster, we use an integral scheduling system that considers both Space and Time sharing concerns. For dealing with the Time Sharing (TS) aspect, we use a technique based on the communication-driven coscheduling principle. This kind of TS system has some implications on the Space Sharing (SS) system, that force us to modify the way job scheduling is traditionally done. In this paper, we analyze the relation between the TS and the SS systems in a non-dedicated cluster. As a consequence of this analysis, we propose a new technique, termed 3DBackfilling. This proposal implements the well known SS technique of backfilling, but applied to an environment with a MultiProgramming Level (MPL) of the parallel applications that is greater than one. Besides, 3DBackfilling considers the requirements of the local workload running on each node. Our proposal was evaluated in a PVM/MPI Linux cluster, and it was compared with several more traditional SS policies applied to non-dedicated environments
VI Workshop de Procesamiento Distribuido y Paralelo (WPDP)
Red de Universidades con Carreras en Informática (RedUNCI)
Materia
Ciencias Informáticas
Distributed Systems
Scheduling
Space-Sharing,
non-dedicated clusters
backfilling
Nivel de accesibilidad
acceso abierto
Condiciones de uso
http://creativecommons.org/licenses/by-nc-sa/2.5/ar/
Repositorio
SEDICI (UNLP)
Institución
Universidad Nacional de La Plata
OAI Identificador
oai:sedici.unlp.edu.ar:10915/23171

id SEDICI_2c24deadd9939dad4cc0a4770a5c2de1
oai_identifier_str oai:sedici.unlp.edu.ar:10915/23171
network_acronym_str SEDICI
repository_id_str 1329
network_name_str SEDICI (UNLP)
spelling Applying backfilling over a non-dedicated clusterHanzich, MauricioGiné de Solà, FrancescHernández Budé, PorfidioSolsona Theás, FrancescLuque Fadón, EmilioCiencias InformáticasDistributed SystemsSchedulingSpace-Sharing,non-dedicated clustersbackfillingThe resource utilization level in open laboratories of several universities has been shown to be very low. Our aim is to take advantage of those idle resources for parallel computation without disturbing the local load. In order to provide a system that lets us execute parallel applications in such a non-dedicated cluster, we use an integral scheduling system that considers both Space and Time sharing concerns. For dealing with the Time Sharing (TS) aspect, we use a technique based on the communication-driven coscheduling principle. This kind of TS system has some implications on the Space Sharing (SS) system, that force us to modify the way job scheduling is traditionally done. In this paper, we analyze the relation between the TS and the SS systems in a non-dedicated cluster. As a consequence of this analysis, we propose a new technique, termed 3DBackfilling. This proposal implements the well known SS technique of backfilling, but applied to an environment with a MultiProgramming Level (MPL) of the parallel applications that is greater than one. Besides, 3DBackfilling considers the requirements of the local workload running on each node. Our proposal was evaluated in a PVM/MPI Linux cluster, and it was compared with several more traditional SS policies applied to non-dedicated environmentsVI Workshop de Procesamiento Distribuido y Paralelo (WPDP)Red de Universidades con Carreras en Informática (RedUNCI)2005-10info:eu-repo/semantics/conferenceObjectinfo:eu-repo/semantics/publishedVersionObjeto de conferenciahttp://purl.org/coar/resource_type/c_5794info:ar-repo/semantics/documentoDeConferenciaapplication/pdfhttp://sedici.unlp.edu.ar/handle/10915/23171enginfo:eu-repo/semantics/openAccesshttp://creativecommons.org/licenses/by-nc-sa/2.5/ar/Creative Commons Attribution-NonCommercial-ShareAlike 2.5 Argentina (CC BY-NC-SA 2.5)reponame:SEDICI (UNLP)instname:Universidad Nacional de La Platainstacron:UNLP2025-09-29T10:55:21Zoai:sedici.unlp.edu.ar:10915/23171Institucionalhttp://sedici.unlp.edu.ar/Universidad públicaNo correspondehttp://sedici.unlp.edu.ar/oai/snrdalira@sedici.unlp.edu.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:13292025-09-29 10:55:21.818SEDICI (UNLP) - Universidad Nacional de La Platafalse
dc.title.none.fl_str_mv Applying backfilling over a non-dedicated cluster
title Applying backfilling over a non-dedicated cluster
spellingShingle Applying backfilling over a non-dedicated cluster
Hanzich, Mauricio
Ciencias Informáticas
Distributed Systems
Scheduling
Space-Sharing,
non-dedicated clusters
backfilling
title_short Applying backfilling over a non-dedicated cluster
title_full Applying backfilling over a non-dedicated cluster
title_fullStr Applying backfilling over a non-dedicated cluster
title_full_unstemmed Applying backfilling over a non-dedicated cluster
title_sort Applying backfilling over a non-dedicated cluster
dc.creator.none.fl_str_mv Hanzich, Mauricio
Giné de Solà, Francesc
Hernández Budé, Porfidio
Solsona Theás, Francesc
Luque Fadón, Emilio
author Hanzich, Mauricio
author_facet Hanzich, Mauricio
Giné de Solà, Francesc
Hernández Budé, Porfidio
Solsona Theás, Francesc
Luque Fadón, Emilio
author_role author
author2 Giné de Solà, Francesc
Hernández Budé, Porfidio
Solsona Theás, Francesc
Luque Fadón, Emilio
author2_role author
author
author
author
dc.subject.none.fl_str_mv Ciencias Informáticas
Distributed Systems
Scheduling
Space-Sharing,
non-dedicated clusters
backfilling
topic Ciencias Informáticas
Distributed Systems
Scheduling
Space-Sharing,
non-dedicated clusters
backfilling
dc.description.none.fl_txt_mv The resource utilization level in open laboratories of several universities has been shown to be very low. Our aim is to take advantage of those idle resources for parallel computation without disturbing the local load. In order to provide a system that lets us execute parallel applications in such a non-dedicated cluster, we use an integral scheduling system that considers both Space and Time sharing concerns. For dealing with the Time Sharing (TS) aspect, we use a technique based on the communication-driven coscheduling principle. This kind of TS system has some implications on the Space Sharing (SS) system, that force us to modify the way job scheduling is traditionally done. In this paper, we analyze the relation between the TS and the SS systems in a non-dedicated cluster. As a consequence of this analysis, we propose a new technique, termed 3DBackfilling. This proposal implements the well known SS technique of backfilling, but applied to an environment with a MultiProgramming Level (MPL) of the parallel applications that is greater than one. Besides, 3DBackfilling considers the requirements of the local workload running on each node. Our proposal was evaluated in a PVM/MPI Linux cluster, and it was compared with several more traditional SS policies applied to non-dedicated environments
VI Workshop de Procesamiento Distribuido y Paralelo (WPDP)
Red de Universidades con Carreras en Informática (RedUNCI)
description The resource utilization level in open laboratories of several universities has been shown to be very low. Our aim is to take advantage of those idle resources for parallel computation without disturbing the local load. In order to provide a system that lets us execute parallel applications in such a non-dedicated cluster, we use an integral scheduling system that considers both Space and Time sharing concerns. For dealing with the Time Sharing (TS) aspect, we use a technique based on the communication-driven coscheduling principle. This kind of TS system has some implications on the Space Sharing (SS) system, that force us to modify the way job scheduling is traditionally done. In this paper, we analyze the relation between the TS and the SS systems in a non-dedicated cluster. As a consequence of this analysis, we propose a new technique, termed 3DBackfilling. This proposal implements the well known SS technique of backfilling, but applied to an environment with a MultiProgramming Level (MPL) of the parallel applications that is greater than one. Besides, 3DBackfilling considers the requirements of the local workload running on each node. Our proposal was evaluated in a PVM/MPI Linux cluster, and it was compared with several more traditional SS policies applied to non-dedicated environments
publishDate 2005
dc.date.none.fl_str_mv 2005-10
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/10915/23171
url http://sedici.unlp.edu.ar/handle/10915/23171
dc.language.none.fl_str_mv eng
language eng
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
http://creativecommons.org/licenses/by-nc-sa/2.5/ar/
Creative Commons Attribution-NonCommercial-ShareAlike 2.5 Argentina (CC BY-NC-SA 2.5)
eu_rights_str_mv openAccess
rights_invalid_str_mv http://creativecommons.org/licenses/by-nc-sa/2.5/ar/
Creative Commons Attribution-NonCommercial-ShareAlike 2.5 Argentina (CC BY-NC-SA 2.5)
dc.format.none.fl_str_mv application/pdf
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_ 1844615812318494720
score 13.070432