Virtual topologies for massively parallel computations. (c2015)

In their essence, recursive search tree algorithms are nothing but mere enumeration of the solution space. Most of the time is spent generating new search tree nodes with very little computation performed at each tree node. Massively parallel computations that are based on such algorithms often suff...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Jahed, Karim A. (author)
التنسيق: masterThesis
منشور في: 2015
الموضوعات:
الوصول للمادة أونلاين:http://hdl.handle.net/10725/2122
https://doi.org/10.26756/th.2015.13
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513457406083072
author Jahed, Karim A.
author_facet Jahed, Karim A.
author_role author
dc.creator.none.fl_str_mv Jahed, Karim A.
dc.date.none.fl_str_mv 2015-09-09T05:55:58Z
2015-09-09T05:55:58Z
2015
2015-09-09
2015-05-28
dc.identifier.none.fl_str_mv http://hdl.handle.net/10725/2122
https://doi.org/10.26756/th.2015.13
dc.language.none.fl_str_mv en
dc.publisher.none.fl_str_mv Lebanese American University
dc.rights.*.fl_str_mv info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Virtual computer systems
Electric network topology
Parallel processing (Electronic computers)
Lebanese American University -- Dissertations
Dissertations, Academic
dc.title.none.fl_str_mv Virtual topologies for massively parallel computations. (c2015)
a case study
dc.type.none.fl_str_mv Thesis
info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/masterThesis
description In their essence, recursive search tree algorithms are nothing but mere enumeration of the solution space. Most of the time is spent generating new search tree nodes with very little computation performed at each tree node. Massively parallel computations that are based on such algorithms often suffer from a large communication overhead due to the exponential growth in the number of tasks generated at each search-tree level. Cores would spend computation time generating and exchanging tasks rather than traversing the search tree. This communication overhead will only increase as we scale-up the computation, up to the point where adding new cores negatively affects the computational time. To address this issue, we propose virtual topologies: an architecture-oblivious communication graph imposed on top of the physical network to limit and manage core-to-core communication. Using the Cluster Editing problem as a case study, we show that managed cooperation, coupled with an efficient task generation and load balancing strategy, is capable of dramatically reducing the communication overhead and improving the computational throughput.
eu_rights_str_mv openAccess
format masterThesis
id LAURepo_01ba1f207489ee4da3c46fc2933c216c
language_invalid_str_mv en
network_acronym_str LAURepo
network_name_str Lebanese American University repository
oai_identifier_str oai:laur.lau.edu.lb:10725/2122
publishDate 2015
publisher.none.fl_str_mv Lebanese American University
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Virtual topologies for massively parallel computations. (c2015)a case studyJahed, Karim A.Virtual computer systemsElectric network topologyParallel processing (Electronic computers)Lebanese American University -- DissertationsDissertations, AcademicIn their essence, recursive search tree algorithms are nothing but mere enumeration of the solution space. Most of the time is spent generating new search tree nodes with very little computation performed at each tree node. Massively parallel computations that are based on such algorithms often suffer from a large communication overhead due to the exponential growth in the number of tasks generated at each search-tree level. Cores would spend computation time generating and exchanging tasks rather than traversing the search tree. This communication overhead will only increase as we scale-up the computation, up to the point where adding new cores negatively affects the computational time. To address this issue, we propose virtual topologies: an architecture-oblivious communication graph imposed on top of the physical network to limit and manage core-to-core communication. Using the Cluster Editing problem as a case study, we show that managed cooperation, coupled with an efficient task generation and load balancing strategy, is capable of dramatically reducing the communication overhead and improving the computational throughput.N/A1 hard copy: x, 42 leaves; col. ill.; 31 cm. avaiable at RNL.Bibliography: leaves 40-42.Lebanese American University2015-09-09T05:55:58Z2015-09-09T05:55:58Z20152015-09-092015-05-28Thesisinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesishttp://hdl.handle.net/10725/2122https://doi.org/10.26756/th.2015.13eninfo:eu-repo/semantics/openAccessoai:laur.lau.edu.lb:10725/21222020-11-20T08:43:22Z
spellingShingle Virtual topologies for massively parallel computations. (c2015)
Jahed, Karim A.
Virtual computer systems
Electric network topology
Parallel processing (Electronic computers)
Lebanese American University -- Dissertations
Dissertations, Academic
status_str publishedVersion
title Virtual topologies for massively parallel computations. (c2015)
title_full Virtual topologies for massively parallel computations. (c2015)
title_fullStr Virtual topologies for massively parallel computations. (c2015)
title_full_unstemmed Virtual topologies for massively parallel computations. (c2015)
title_short Virtual topologies for massively parallel computations. (c2015)
title_sort Virtual topologies for massively parallel computations. (c2015)
topic Virtual computer systems
Electric network topology
Parallel processing (Electronic computers)
Lebanese American University -- Dissertations
Dissertations, Academic
url http://hdl.handle.net/10725/2122
https://doi.org/10.26756/th.2015.13