An improved evolutlonary approach for the cluster allocation problem

Autores
Apolloni, Rubén; Molina, Silvia; Gallard, Raúl Hector
Año de publicación
1999
Idioma
inglés
Tipo de recurso
documento de conferencia
Estado
versión publicada
Descripción
In a distributed system, consisting of a set of interconnected local area networks, users migrate to different machines, users invoke different programs and users and programs need distinct data files to satisfy their expectations. Consequently optimal allocation of parallel program tasks can increase system performance as results of traffic cost reducti9n between clusters2• The problem of allocating a program in a particular system node can be divided into two subproblems: i) allocate the program in a cluster such that traffic costs are minimized and ii) within a particular cluster choose the node following sorne load balancing criteria [5]. To solve subproblem i), in 1992 U. M. Borghoff [2] proposed the Individual Program Execution Location Algorithm IPELA, where essentially giving a distribution of data files the best allocation for program execution, minimizing the expected intercluster traffic, is searched. The algorithm uses diverse input data such us the cost for starting a program at sorne node [10], thedependencies between program and data files [1], separated read and write access costs [9], the impact of l/O activities on the communication costs [8] and the allocation of program and data files [3]. . As the number of possible allocations induce high complexity and the model could not be solved too optimality Borghoff reduced the number of combinations by limiting the number of data file replicas and looking for those combinations where the relevant file sets's allocation is varied. This approach reduced complexity. Nevertheless running ¡PELA implied evaluation of each solution in a large problem space.
Eje: Redes y sistemas inteligentes
Red de Universidades con Carreras en Informática (RedUNCI)
Materia
Ciencias Informáticas
ARTIFICIAL INTELLIGENCE
cluster allocation problem
Clustering
improved evolutlonary
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:10915/22226

id SEDICI_c81c9ab770e9a7df1abb926a92c0dafd
oai_identifier_str oai:sedici.unlp.edu.ar:10915/22226
network_acronym_str SEDICI
repository_id_str 1329
network_name_str SEDICI (UNLP)
spelling An improved evolutlonary approach for the cluster allocation problemApolloni, RubénMolina, SilviaGallard, Raúl HectorCiencias InformáticasARTIFICIAL INTELLIGENCEcluster allocation problemClusteringimproved evolutlonaryIn a distributed system, consisting of a set of interconnected local area networks, users migrate to different machines, users invoke different programs and users and programs need distinct data files to satisfy their expectations. Consequently optimal allocation of parallel program tasks can increase system performance as results of traffic cost reducti9n between clusters2• The problem of allocating a program in a particular system node can be divided into two subproblems: i) allocate the program in a cluster such that traffic costs are minimized and ii) within a particular cluster choose the node following sorne load balancing criteria [5]. To solve subproblem i), in 1992 U. M. Borghoff [2] proposed the Individual Program Execution Location Algorithm IPELA, where essentially giving a distribution of data files the best allocation for program execution, minimizing the expected intercluster traffic, is searched. The algorithm uses diverse input data such us the cost for starting a program at sorne node [10], thedependencies between program and data files [1], separated read and write access costs [9], the impact of l/O activities on the communication costs [8] and the allocation of program and data files [3]. . As the number of possible allocations induce high complexity and the model could not be solved too optimality Borghoff reduced the number of combinations by limiting the number of data file replicas and looking for those combinations where the relevant file sets's allocation is varied. This approach reduced complexity. Nevertheless running ¡PELA implied evaluation of each solution in a large problem space.Eje: Redes y sistemas inteligentesRed de Universidades con Carreras en Informática (RedUNCI)1999-05info: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/10915/22226enginfo: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:UNLP2025-09-29T10:54:57Zoai:sedici.unlp.edu.ar:10915/22226Institucionalhttp://sedici.unlp.edu.ar/Universidad públicaNo correspondehttp://sedici.unlp.edu.ar/oai/snrdalira@sedici.unlp.edu.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:13292025-09-29 10:54:58.045SEDICI (UNLP) - Universidad Nacional de La Platafalse
dc.title.none.fl_str_mv An improved evolutlonary approach for the cluster allocation problem
title An improved evolutlonary approach for the cluster allocation problem
spellingShingle An improved evolutlonary approach for the cluster allocation problem
Apolloni, Rubén
Ciencias Informáticas
ARTIFICIAL INTELLIGENCE
cluster allocation problem
Clustering
improved evolutlonary
title_short An improved evolutlonary approach for the cluster allocation problem
title_full An improved evolutlonary approach for the cluster allocation problem
title_fullStr An improved evolutlonary approach for the cluster allocation problem
title_full_unstemmed An improved evolutlonary approach for the cluster allocation problem
title_sort An improved evolutlonary approach for the cluster allocation problem
dc.creator.none.fl_str_mv Apolloni, Rubén
Molina, Silvia
Gallard, Raúl Hector
author Apolloni, Rubén
author_facet Apolloni, Rubén
Molina, Silvia
Gallard, Raúl Hector
author_role author
author2 Molina, Silvia
Gallard, Raúl Hector
author2_role author
author
dc.subject.none.fl_str_mv Ciencias Informáticas
ARTIFICIAL INTELLIGENCE
cluster allocation problem
Clustering
improved evolutlonary
topic Ciencias Informáticas
ARTIFICIAL INTELLIGENCE
cluster allocation problem
Clustering
improved evolutlonary
dc.description.none.fl_txt_mv In a distributed system, consisting of a set of interconnected local area networks, users migrate to different machines, users invoke different programs and users and programs need distinct data files to satisfy their expectations. Consequently optimal allocation of parallel program tasks can increase system performance as results of traffic cost reducti9n between clusters2• The problem of allocating a program in a particular system node can be divided into two subproblems: i) allocate the program in a cluster such that traffic costs are minimized and ii) within a particular cluster choose the node following sorne load balancing criteria [5]. To solve subproblem i), in 1992 U. M. Borghoff [2] proposed the Individual Program Execution Location Algorithm IPELA, where essentially giving a distribution of data files the best allocation for program execution, minimizing the expected intercluster traffic, is searched. The algorithm uses diverse input data such us the cost for starting a program at sorne node [10], thedependencies between program and data files [1], separated read and write access costs [9], the impact of l/O activities on the communication costs [8] and the allocation of program and data files [3]. . As the number of possible allocations induce high complexity and the model could not be solved too optimality Borghoff reduced the number of combinations by limiting the number of data file replicas and looking for those combinations where the relevant file sets's allocation is varied. This approach reduced complexity. Nevertheless running ¡PELA implied evaluation of each solution in a large problem space.
Eje: Redes y sistemas inteligentes
Red de Universidades con Carreras en Informática (RedUNCI)
description In a distributed system, consisting of a set of interconnected local area networks, users migrate to different machines, users invoke different programs and users and programs need distinct data files to satisfy their expectations. Consequently optimal allocation of parallel program tasks can increase system performance as results of traffic cost reducti9n between clusters2• The problem of allocating a program in a particular system node can be divided into two subproblems: i) allocate the program in a cluster such that traffic costs are minimized and ii) within a particular cluster choose the node following sorne load balancing criteria [5]. To solve subproblem i), in 1992 U. M. Borghoff [2] proposed the Individual Program Execution Location Algorithm IPELA, where essentially giving a distribution of data files the best allocation for program execution, minimizing the expected intercluster traffic, is searched. The algorithm uses diverse input data such us the cost for starting a program at sorne node [10], thedependencies between program and data files [1], separated read and write access costs [9], the impact of l/O activities on the communication costs [8] and the allocation of program and data files [3]. . As the number of possible allocations induce high complexity and the model could not be solved too optimality Borghoff reduced the number of combinations by limiting the number of data file replicas and looking for those combinations where the relevant file sets's allocation is varied. This approach reduced complexity. Nevertheless running ¡PELA implied evaluation of each solution in a large problem space.
publishDate 1999
dc.date.none.fl_str_mv 1999-05
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/10915/22226
url http://sedici.unlp.edu.ar/handle/10915/22226
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_ 1844615807874629632
score 13.070432