Impact Assessment on the Parallel Performance of Node-Core Combinations in a Multicore Cluster Environment: A Case of Study

Multicore processors have opened new paths for improving the parallel performance in cluster environments. Nevertheless, the selection of different combinations between the amount of nodes and the number of cores per node implies different results in terms of parallel performance. We performed an im...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Fernández, César, Saravia, Francisco, Valle, Carlos, Allende, Héctor
Formato: Objeto de conferencia
Lenguaje:Inglés
Publicado: 2010
Materias:
Ciencias Informáticas
Parallel Algorithms
Parallelism and Data Sharing on Multicore Architectures
Ensemble Learning
Local Negative Correlation
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/152736
http://39jaiio.sadio.org.ar/sites/default/files/39jaiio-hpc-17.pdf
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-152736
record_format dspace
spelling I19-R120-10915-1527362023-05-10T20:02:46Z http://sedici.unlp.edu.ar/handle/10915/152736 http://39jaiio.sadio.org.ar/sites/default/files/39jaiio-hpc-17.pdf issn:1851-9326 Impact Assessment on the Parallel Performance of Node-Core Combinations in a Multicore Cluster Environment: A Case of Study Fernández, César Saravia, Francisco Valle, Carlos Allende, Héctor 2010 2010 2023-05-10T17:16:15Z en Ciencias Informáticas Parallel Algorithms Parallelism and Data Sharing on Multicore Architectures Ensemble Learning Local Negative Correlation Multicore processors have opened new paths for improving the parallel performance in cluster environments. Nevertheless, the selection of different combinations between the amount of nodes and the number of cores per node implies different results in terms of parallel performance. We performed an impact assessment on the parallel performance of node-core combinations using a parallel approach of a machine learning ensemble algorithm. Our results reveal that two key factors for selecting a suitable node-core combination: the network capabilities and the workload distribution. We observed that the network interconnection limits the amount of nodes that can be efficiently used, due to the extranode communications does not allow to keep scaling as the number of nodes is increased. The best results were obtained by reaching a balance between intra-node and extra-node communications. By the other hand, the parallel performance can be negatively affected when the workload distribution is not homogeneous among nodes. Sociedad Argentina de Informática e Investigación Operativa Objeto de conferencia Objeto de conferencia http://creativecommons.org/licenses/by-nc-sa/4.0/ Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) application/pdf 3363-3378
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
Parallel Algorithms
Parallelism and Data Sharing on Multicore Architectures
Ensemble Learning
Local Negative Correlation
spellingShingle Ciencias Informáticas
Parallel Algorithms
Parallelism and Data Sharing on Multicore Architectures
Ensemble Learning
Local Negative Correlation
Fernández, César
Saravia, Francisco
Valle, Carlos
Allende, Héctor
Impact Assessment on the Parallel Performance of Node-Core Combinations in a Multicore Cluster Environment: A Case of Study
topic_facet Ciencias Informáticas
Parallel Algorithms
Parallelism and Data Sharing on Multicore Architectures
Ensemble Learning
Local Negative Correlation
description Multicore processors have opened new paths for improving the parallel performance in cluster environments. Nevertheless, the selection of different combinations between the amount of nodes and the number of cores per node implies different results in terms of parallel performance. We performed an impact assessment on the parallel performance of node-core combinations using a parallel approach of a machine learning ensemble algorithm. Our results reveal that two key factors for selecting a suitable node-core combination: the network capabilities and the workload distribution. We observed that the network interconnection limits the amount of nodes that can be efficiently used, due to the extranode communications does not allow to keep scaling as the number of nodes is increased. The best results were obtained by reaching a balance between intra-node and extra-node communications. By the other hand, the parallel performance can be negatively affected when the workload distribution is not homogeneous among nodes.
format Objeto de conferencia
Objeto de conferencia
author Fernández, César
Saravia, Francisco
Valle, Carlos
Allende, Héctor
author_facet Fernández, César
Saravia, Francisco
Valle, Carlos
Allende, Héctor
author_sort Fernández, César
title Impact Assessment on the Parallel Performance of Node-Core Combinations in a Multicore Cluster Environment: A Case of Study
title_short Impact Assessment on the Parallel Performance of Node-Core Combinations in a Multicore Cluster Environment: A Case of Study
title_full Impact Assessment on the Parallel Performance of Node-Core Combinations in a Multicore Cluster Environment: A Case of Study
title_fullStr Impact Assessment on the Parallel Performance of Node-Core Combinations in a Multicore Cluster Environment: A Case of Study
title_full_unstemmed Impact Assessment on the Parallel Performance of Node-Core Combinations in a Multicore Cluster Environment: A Case of Study
title_sort impact assessment on the parallel performance of node-core combinations in a multicore cluster environment: a case of study
publishDate 2010
url http://sedici.unlp.edu.ar/handle/10915/152736
http://39jaiio.sadio.org.ar/sites/default/files/39jaiio-hpc-17.pdf
work_keys_str_mv AT fernandezcesar impactassessmentontheparallelperformanceofnodecorecombinationsinamulticoreclusterenvironmentacaseofstudy
AT saraviafrancisco impactassessmentontheparallelperformanceofnodecorecombinationsinamulticoreclusterenvironmentacaseofstudy
AT vallecarlos impactassessmentontheparallelperformanceofnodecorecombinationsinamulticoreclusterenvironmentacaseofstudy
AT allendehector impactassessmentontheparallelperformanceofnodecorecombinationsinamulticoreclusterenvironmentacaseofstudy
_version_ 1765660143463170048

Ejemplares similares