Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computations

Three optimization methods derived from natural sciences are considered for allocating data to multicomputer nodes. These are simulated annealing, genetic algorithms and neural networks. A number of design choices and the addition of preprocessing and postprocessing steps lead to versions of the alg...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Mansour, Nashat (author)
مؤلفون آخرون: Fox, Geoffrey C. (author)
التنسيق: article
منشور في: 1992
الوصول للمادة أونلاين:http://hdl.handle.net/10725/2945
http://dx.doi.org/10.1002/cpe.4330040705
http://onlinelibrary.wiley.com/doi/10.1002/cpe.4330040705/full
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513459487506432
author Mansour, Nashat
author2 Fox, Geoffrey C.
author2_role author
author_facet Mansour, Nashat
Fox, Geoffrey C.
author_role author
dc.creator.none.fl_str_mv Mansour, Nashat
Fox, Geoffrey C.
dc.date.none.fl_str_mv 1992
2016-01-25T12:46:37Z
2016-01-25T12:46:37Z
2016-01-25
dc.identifier.none.fl_str_mv 1532-0626
http://hdl.handle.net/10725/2945
http://dx.doi.org/10.1002/cpe.4330040705
Mansour, N., & Fox, G. C. (1992). Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computations. Concurrency: practice and experience, 4(7), 557-574.
http://onlinelibrary.wiley.com/doi/10.1002/cpe.4330040705/full
dc.language.none.fl_str_mv en
dc.relation.none.fl_str_mv Concurrency and computation
dc.rights.*.fl_str_mv info:eu-repo/semantics/openAccess
dc.title.none.fl_str_mv Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computations
dc.type.none.fl_str_mv Article
info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/article
description Three optimization methods derived from natural sciences are considered for allocating data to multicomputer nodes. These are simulated annealing, genetic algorithms and neural networks. A number of design choices and the addition of preprocessing and postprocessing steps lead to versions of the algorithms which differ in solution qualities and execution times. In this paper the performances of these versions are critically evaluated and compared for test cases with different features. The performance criteria are solution quality, execution time, robustness, bias and parallelizability. Experimental results show that the physical algorithms produce better solutions than those of recursive bisection methods and that they have diverse properties. Hence, different algorithms would be suitable for different applications. For example, the annealing and genetic algorithms produce better solutions and do not show a bias towards particular problem structures, but they are slower than the neural network algorithms. Preprocessing graph contraction is one of the additional steps suggested for the physical methods. It produces a significant reduction in execution time, which is necessary for their applicability to large problems.
eu_rights_str_mv openAccess
format article
id LAURepo_e6ec34d0e8c5382bb075a8e00d65d4a1
identifier_str_mv 1532-0626
Mansour, N., & Fox, G. C. (1992). Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computations. Concurrency: practice and experience, 4(7), 557-574.
language_invalid_str_mv en
network_acronym_str LAURepo
network_name_str Lebanese American University repository
oai_identifier_str oai:laur.lau.edu.lb:10725/2945
publishDate 1992
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computationsMansour, NashatFox, Geoffrey C.Three optimization methods derived from natural sciences are considered for allocating data to multicomputer nodes. These are simulated annealing, genetic algorithms and neural networks. A number of design choices and the addition of preprocessing and postprocessing steps lead to versions of the algorithms which differ in solution qualities and execution times. In this paper the performances of these versions are critically evaluated and compared for test cases with different features. The performance criteria are solution quality, execution time, robustness, bias and parallelizability. Experimental results show that the physical algorithms produce better solutions than those of recursive bisection methods and that they have diverse properties. Hence, different algorithms would be suitable for different applications. For example, the annealing and genetic algorithms produce better solutions and do not show a bias towards particular problem structures, but they are slower than the neural network algorithms. Preprocessing graph contraction is one of the additional steps suggested for the physical methods. It produces a significant reduction in execution time, which is necessary for their applicability to large problems.PublishedN/A2016-01-25T12:46:37Z2016-01-25T12:46:37Z19922016-01-25Articleinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/article1532-0626http://hdl.handle.net/10725/2945http://dx.doi.org/10.1002/cpe.4330040705Mansour, N., & Fox, G. C. (1992). Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computations. Concurrency: practice and experience, 4(7), 557-574.http://onlinelibrary.wiley.com/doi/10.1002/cpe.4330040705/fullenConcurrency and computationinfo:eu-repo/semantics/openAccessoai:laur.lau.edu.lb:10725/29452017-04-07T12:22:50Z
spellingShingle Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computations
Mansour, Nashat
status_str publishedVersion
title Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computations
title_full Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computations
title_fullStr Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computations
title_full_unstemmed Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computations
title_short Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computations
title_sort Allocating data to multicomputer nodes by physical optimization algorithms for loosely synchronous computations
url http://hdl.handle.net/10725/2945
http://dx.doi.org/10.1002/cpe.4330040705
http://onlinelibrary.wiley.com/doi/10.1002/cpe.4330040705/full