Experimental research of a shared memory subsystem with limited queue length for specialized reconfigurable multiprocessor systems
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Independent Journal of Management & Production |
Texto Completo: | http://www.ijmp.jor.br/index.php/ijmp/article/view/1922 |
Resumo: | Recently, reconfigurable systems based on field programmable logic devices (FPLDs) have been widely used in high-performance computing. The paper discusses issues related to the experimental research of a shared memory subsystem with a limited queue length of specialized reconfigurable multiprocessor systems using the developed mathematical modelling method. The paper presents the results of the method proposed by the authors for modelling multiprocessor systems based on open queuing networks with limited queue lengths. Based on these conditions, as well as the architectural features of the investigated processor-memory subsystem, expressions are calculated to estimate the exchange time and the resulting delays at each exchange stage. During the research, the main attention was paid to the dependence of the increase in the number of processor nodes in the processor-memory subsystem. As a result, the data obtained showed that the processor growth significantly affects the exchange time, creating a significant load on the common bus, as well as increasing delays at the stages when request transfer operation from the processor to the memory is performed. At the same time, the inadequate behaviour of experimental results and inaccuracy of their values when using the basic modelling method are explicitly tracked, which is reflected in the obtained graphs. Computational experiments were carried out to calculate the probabilistic-temporal characteristics of the "processor-memory" subsystem using the developed mathematical modelling methods. Based on the experimental results, it was determined that the delays occurring in subsystem's nodes and the time of exchange between the processor and memory modules depend on the query parameters and the processor-memory subsystem’s architectural characteristics. |
id |
IJMP_005784466c9b2406995ed1af73e2063c |
---|---|
oai_identifier_str |
oai:www.ijmp.jor.br:article/1922 |
network_acronym_str |
IJMP |
network_name_str |
Independent Journal of Management & Production |
repository_id_str |
|
spelling |
Experimental research of a shared memory subsystem with limited queue length for specialized reconfigurable multiprocessor systemsMultiprocessor systemmathematical modellingcomputational experimentprocessor-memory subsystemmemory architectureRecently, reconfigurable systems based on field programmable logic devices (FPLDs) have been widely used in high-performance computing. The paper discusses issues related to the experimental research of a shared memory subsystem with a limited queue length of specialized reconfigurable multiprocessor systems using the developed mathematical modelling method. The paper presents the results of the method proposed by the authors for modelling multiprocessor systems based on open queuing networks with limited queue lengths. Based on these conditions, as well as the architectural features of the investigated processor-memory subsystem, expressions are calculated to estimate the exchange time and the resulting delays at each exchange stage. During the research, the main attention was paid to the dependence of the increase in the number of processor nodes in the processor-memory subsystem. As a result, the data obtained showed that the processor growth significantly affects the exchange time, creating a significant load on the common bus, as well as increasing delays at the stages when request transfer operation from the processor to the memory is performed. At the same time, the inadequate behaviour of experimental results and inaccuracy of their values when using the basic modelling method are explicitly tracked, which is reflected in the obtained graphs. Computational experiments were carried out to calculate the probabilistic-temporal characteristics of the "processor-memory" subsystem using the developed mathematical modelling methods. Based on the experimental results, it was determined that the delays occurring in subsystem's nodes and the time of exchange between the processor and memory modules depend on the query parameters and the processor-memory subsystem’s architectural characteristics.Independent2022-06-01info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttp://www.ijmp.jor.br/index.php/ijmp/article/view/192210.14807/ijmp.v13i4.1922Independent Journal of Management & Production; Vol. 13 No. 4 (2022): Independent Journal of Management & Production (Special Edition CIMEE); s415-s4242236-269X2236-269Xreponame:Independent Journal of Management & Productioninstname:Instituto Federal de Educação, Ciência e Tecnologia de São Paulo (IFSP)instacron:IJM&Penghttp://www.ijmp.jor.br/index.php/ijmp/article/view/1922/2153Copyright (c) 2022 Alexey I. Martyshkin, Dmitry S. Martens-Atyushevhttp://creativecommons.org/licenses/by-nc-sa/4.0info:eu-repo/semantics/openAccessMartyshkin, Alexey I.Martens-Atyushev, Dmitry S.2022-06-01T13:45:13Zoai:www.ijmp.jor.br:article/1922Revistahttp://www.ijmp.jor.br/PUBhttp://www.ijmp.jor.br/index.php/ijmp/oaiijmp@ijmp.jor.br||paulo@paulorodrigues.pro.br||2236-269X2236-269Xopendoar:2022-06-01T13:45:13Independent Journal of Management & Production - Instituto Federal de Educação, Ciência e Tecnologia de São Paulo (IFSP)false |
dc.title.none.fl_str_mv |
Experimental research of a shared memory subsystem with limited queue length for specialized reconfigurable multiprocessor systems |
title |
Experimental research of a shared memory subsystem with limited queue length for specialized reconfigurable multiprocessor systems |
spellingShingle |
Experimental research of a shared memory subsystem with limited queue length for specialized reconfigurable multiprocessor systems Martyshkin, Alexey I. Multiprocessor system mathematical modelling computational experiment processor-memory subsystem memory architecture |
title_short |
Experimental research of a shared memory subsystem with limited queue length for specialized reconfigurable multiprocessor systems |
title_full |
Experimental research of a shared memory subsystem with limited queue length for specialized reconfigurable multiprocessor systems |
title_fullStr |
Experimental research of a shared memory subsystem with limited queue length for specialized reconfigurable multiprocessor systems |
title_full_unstemmed |
Experimental research of a shared memory subsystem with limited queue length for specialized reconfigurable multiprocessor systems |
title_sort |
Experimental research of a shared memory subsystem with limited queue length for specialized reconfigurable multiprocessor systems |
author |
Martyshkin, Alexey I. |
author_facet |
Martyshkin, Alexey I. Martens-Atyushev, Dmitry S. |
author_role |
author |
author2 |
Martens-Atyushev, Dmitry S. |
author2_role |
author |
dc.contributor.author.fl_str_mv |
Martyshkin, Alexey I. Martens-Atyushev, Dmitry S. |
dc.subject.por.fl_str_mv |
Multiprocessor system mathematical modelling computational experiment processor-memory subsystem memory architecture |
topic |
Multiprocessor system mathematical modelling computational experiment processor-memory subsystem memory architecture |
description |
Recently, reconfigurable systems based on field programmable logic devices (FPLDs) have been widely used in high-performance computing. The paper discusses issues related to the experimental research of a shared memory subsystem with a limited queue length of specialized reconfigurable multiprocessor systems using the developed mathematical modelling method. The paper presents the results of the method proposed by the authors for modelling multiprocessor systems based on open queuing networks with limited queue lengths. Based on these conditions, as well as the architectural features of the investigated processor-memory subsystem, expressions are calculated to estimate the exchange time and the resulting delays at each exchange stage. During the research, the main attention was paid to the dependence of the increase in the number of processor nodes in the processor-memory subsystem. As a result, the data obtained showed that the processor growth significantly affects the exchange time, creating a significant load on the common bus, as well as increasing delays at the stages when request transfer operation from the processor to the memory is performed. At the same time, the inadequate behaviour of experimental results and inaccuracy of their values when using the basic modelling method are explicitly tracked, which is reflected in the obtained graphs. Computational experiments were carried out to calculate the probabilistic-temporal characteristics of the "processor-memory" subsystem using the developed mathematical modelling methods. Based on the experimental results, it was determined that the delays occurring in subsystem's nodes and the time of exchange between the processor and memory modules depend on the query parameters and the processor-memory subsystem’s architectural characteristics. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-06-01 |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://www.ijmp.jor.br/index.php/ijmp/article/view/1922 10.14807/ijmp.v13i4.1922 |
url |
http://www.ijmp.jor.br/index.php/ijmp/article/view/1922 |
identifier_str_mv |
10.14807/ijmp.v13i4.1922 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
http://www.ijmp.jor.br/index.php/ijmp/article/view/1922/2153 |
dc.rights.driver.fl_str_mv |
Copyright (c) 2022 Alexey I. Martyshkin, Dmitry S. Martens-Atyushev http://creativecommons.org/licenses/by-nc-sa/4.0 info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Copyright (c) 2022 Alexey I. Martyshkin, Dmitry S. Martens-Atyushev http://creativecommons.org/licenses/by-nc-sa/4.0 |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Independent |
publisher.none.fl_str_mv |
Independent |
dc.source.none.fl_str_mv |
Independent Journal of Management & Production; Vol. 13 No. 4 (2022): Independent Journal of Management & Production (Special Edition CIMEE); s415-s424 2236-269X 2236-269X reponame:Independent Journal of Management & Production instname:Instituto Federal de Educação, Ciência e Tecnologia de São Paulo (IFSP) instacron:IJM&P |
instname_str |
Instituto Federal de Educação, Ciência e Tecnologia de São Paulo (IFSP) |
instacron_str |
IJM&P |
institution |
IJM&P |
reponame_str |
Independent Journal of Management & Production |
collection |
Independent Journal of Management & Production |
repository.name.fl_str_mv |
Independent Journal of Management & Production - Instituto Federal de Educação, Ciência e Tecnologia de São Paulo (IFSP) |
repository.mail.fl_str_mv |
ijmp@ijmp.jor.br||paulo@paulorodrigues.pro.br|| |
_version_ |
1797220494409203712 |