Hopes and facts in evaluating the performance of HPC-I/O on a cloud environment

Currently, there is an increasing interest about the cloud platform by the High Performance Computing (HPC) community, and the Parallel I/O for High Performance Systems is not an exception. In cloud platforms, the user takes into account not only the execution time but also the cost, because the cos...

Full description

Bibliographic Details
Main Authors: Pilar Gómez Sánchez, Sandra Méndez, Dolores Rexachs del Rosario, Emilio Luque
Format: Article
Language:English
Published: Postgraduate Office, School of Computer Science, Universidad Nacional de La Plata 2015-04-01
Series:Journal of Computer Science and Technology
Subjects:
Online Access:https://journal.info.unlp.edu.ar/JCST/article/view/528
id doaj-e92a6072d4e540af986f63effaa81b2d
record_format Article
spelling doaj-e92a6072d4e540af986f63effaa81b2d2021-05-05T13:43:26ZengPostgraduate Office, School of Computer Science, Universidad Nacional de La PlataJournal of Computer Science and Technology1666-60461666-60382015-04-0115012329257Hopes and facts in evaluating the performance of HPC-I/O on a cloud environmentPilar Gómez Sánchez0Sandra Méndez1Dolores Rexachs del RosarioEmilio Luque2Computer Architecture and Operating Systems Department (CAOS), Universität Autónoma de Barcelona, Bellaterra (Barcelona), SpainHigh Performance Systems Division Leibniz Supercomputing Centre (LRZ), Garching (Munich), Germany.Computer Architecture and Operating Systems Department (CAOS), Universität Autónoma de Barcelona, Bellaterra (Barcelona), SpainCurrently, there is an increasing interest about the cloud platform by the High Performance Computing (HPC) community, and the Parallel I/O for High Performance Systems is not an exception. In cloud platforms, the user takes into account not only the execution time but also the cost, because the cost can be one of the most important issue. In this paper, we propose a methodology to quickly evaluate the performance and cost of Virtual Clusters for parallel scientific application that uses parallel I/O. From the parallel application I/O model automatically extracted with our tool PAS2P-IO, we obtain the I/O requirements and then the user can select the Virtual Cluster that meets the application requirements. The application I/O model does not depend on the underlying I/O system. One of the main benefits of applying our methodology is that it is not necessary to execute the application to select the Virtual Cluster on cloud. Finally, costs and performance-cost ratio for the Virtual Clusters are provided to facilitate the decision making on the selection of resources on a cloud platform.https://journal.info.unlp.edu.ar/JCST/article/view/528application i/o modeli/o systemcloud clusteri/o phasesi/o access patterni/o configuration
collection DOAJ
language English
format Article
sources DOAJ
author Pilar Gómez Sánchez
Sandra Méndez
Dolores Rexachs del Rosario
Emilio Luque
spellingShingle Pilar Gómez Sánchez
Sandra Méndez
Dolores Rexachs del Rosario
Emilio Luque
Hopes and facts in evaluating the performance of HPC-I/O on a cloud environment
Journal of Computer Science and Technology
application i/o model
i/o system
cloud cluster
i/o phases
i/o access pattern
i/o configuration
author_facet Pilar Gómez Sánchez
Sandra Méndez
Dolores Rexachs del Rosario
Emilio Luque
author_sort Pilar Gómez Sánchez
title Hopes and facts in evaluating the performance of HPC-I/O on a cloud environment
title_short Hopes and facts in evaluating the performance of HPC-I/O on a cloud environment
title_full Hopes and facts in evaluating the performance of HPC-I/O on a cloud environment
title_fullStr Hopes and facts in evaluating the performance of HPC-I/O on a cloud environment
title_full_unstemmed Hopes and facts in evaluating the performance of HPC-I/O on a cloud environment
title_sort hopes and facts in evaluating the performance of hpc-i/o on a cloud environment
publisher Postgraduate Office, School of Computer Science, Universidad Nacional de La Plata
series Journal of Computer Science and Technology
issn 1666-6046
1666-6038
publishDate 2015-04-01
description Currently, there is an increasing interest about the cloud platform by the High Performance Computing (HPC) community, and the Parallel I/O for High Performance Systems is not an exception. In cloud platforms, the user takes into account not only the execution time but also the cost, because the cost can be one of the most important issue. In this paper, we propose a methodology to quickly evaluate the performance and cost of Virtual Clusters for parallel scientific application that uses parallel I/O. From the parallel application I/O model automatically extracted with our tool PAS2P-IO, we obtain the I/O requirements and then the user can select the Virtual Cluster that meets the application requirements. The application I/O model does not depend on the underlying I/O system. One of the main benefits of applying our methodology is that it is not necessary to execute the application to select the Virtual Cluster on cloud. Finally, costs and performance-cost ratio for the Virtual Clusters are provided to facilitate the decision making on the selection of resources on a cloud platform.
topic application i/o model
i/o system
cloud cluster
i/o phases
i/o access pattern
i/o configuration
url https://journal.info.unlp.edu.ar/JCST/article/view/528
work_keys_str_mv AT pilargomezsanchez hopesandfactsinevaluatingtheperformanceofhpcioonacloudenvironment
AT sandramendez hopesandfactsinevaluatingtheperformanceofhpcioonacloudenvironment
AT doloresrexachsdelrosario hopesandfactsinevaluatingtheperformanceofhpcioonacloudenvironment
AT emilioluque hopesandfactsinevaluatingtheperformanceofhpcioonacloudenvironment
_version_ 1721461480073723904