Using portable monitoring for heterogeneous clusters on Windows and Linux operating systems

This paper describes the advances obtained with the XPVM-W95 2.0, a novel monitoring tool for parallel applications that employ PVM-W95 (PVM for Windows) as well as PVM for Linux. The tool provides, at runtime, the appropriate information about parallel virtual machine configuration, parallel applic...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Kikuti, Daniel, Souza, Paulo S. L., Souza, Simone R. S.
Formato: Articulo
Lenguaje:Inglés
Publicado: 2003
Materias:
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/9462
http://journal.info.unlp.edu.ar/wp-content/uploads/JCST-Oct03-3.pdf
Aporte de:
id I19-R120-10915-9462
record_format dspace
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
Parallel processing
OPERATING SYSTEMS
monitoring
spellingShingle Ciencias Informáticas
Parallel processing
OPERATING SYSTEMS
monitoring
Kikuti, Daniel
Souza, Paulo S. L.
Souza, Simone R. S.
Using portable monitoring for heterogeneous clusters on Windows and Linux operating systems
topic_facet Ciencias Informáticas
Parallel processing
OPERATING SYSTEMS
monitoring
description This paper describes the advances obtained with the XPVM-W95 2.0, a novel monitoring tool for parallel applications that employ PVM-W95 (PVM for Windows) as well as PVM for Linux. The tool provides, at runtime, the appropriate information about parallel virtual machine configuration, parallel applications and workload from each node. The three more important aspects of the XPVM-W95 are: friendly graphical interface, portability and ability to deal with heterogeneity. These items were improved by version 2, mainly when considering the modularity rearrangement. Experiments demonstrate that XPVM-W95 has a stable behavior and reached the objectives proposed. XPVM-W95 allows a great portability of its source code and also allows the monitoring using different metrics. Empirical studies, realized by means of one single application, demonstrated an intrusion of 17,0% on Windows and 0,13% on Linux.
format Articulo
Articulo
author Kikuti, Daniel
Souza, Paulo S. L.
Souza, Simone R. S.
author_facet Kikuti, Daniel
Souza, Paulo S. L.
Souza, Simone R. S.
author_sort Kikuti, Daniel
title Using portable monitoring for heterogeneous clusters on Windows and Linux operating systems
title_short Using portable monitoring for heterogeneous clusters on Windows and Linux operating systems
title_full Using portable monitoring for heterogeneous clusters on Windows and Linux operating systems
title_fullStr Using portable monitoring for heterogeneous clusters on Windows and Linux operating systems
title_full_unstemmed Using portable monitoring for heterogeneous clusters on Windows and Linux operating systems
title_sort using portable monitoring for heterogeneous clusters on windows and linux operating systems
publishDate 2003
url http://sedici.unlp.edu.ar/handle/10915/9462
http://journal.info.unlp.edu.ar/wp-content/uploads/JCST-Oct03-3.pdf
work_keys_str_mv AT kikutidaniel usingportablemonitoringforheterogeneousclustersonwindowsandlinuxoperatingsystems
AT souzapaulosl usingportablemonitoringforheterogeneousclustersonwindowsandlinuxoperatingsystems
AT souzasimoners usingportablemonitoringforheterogeneousclustersonwindowsandlinuxoperatingsystems
bdutipo_str Repositorios
_version_ 1764820491707613185