Platform Calibration for Load Balancing of Large Simulations: TLM Case

Cristian Ruiz; Mihai Alexandru; Olivier Richard; Thierry Monteil; Hervé Aubert

doi:10.1109/CCGrid.2014.26

Platform Calibration for Load Balancing of Large Simulations: TLM Case

Cristian Ruiz ^{1, *} Mihai Alexandru ^{2, 3} Olivier Richard ¹ Thierry Monteil ^{3, 2} Hervé Aubert ^{2, 3}

* Auteur correspondant

1 MESCAL - Middleware efficiently scalable

Inria Grenoble - Rhône-Alpes, LIG - Laboratoire d'Informatique de Grenoble

2 LAAS-SARA - Équipe Services et Architectures pour Réseaux Avancés

LAAS - Laboratoire d'analyse et d'architecture des systèmes [Toulouse]

3 INSA Toulouse - Institut National des Sciences Appliquées - Toulouse

Abstract : The heterogeneous nature of distributed platforms such as computational Grids is one of the main barriers to effectively deploy tightly-coupled applications. For those applications , one common problem that appears due to the hardware heterogeneity is the load imbalance which slows down the application to the pace of the slower processor. One solution is to distribute the load adequately taking into account hardware capacities. To do so, an estimation of the hardware capacities for running the application has to be obtained. In this paper, we present a static load balancing for iterative tightly-coupled applications based on a profile prediction model. This technique is presented as a successful example of the interaction between experiment management tools and parallel applications. The experiment management tool Expo is used that enabled to: (1) provide a general, lightweight and descriptive way to capture the tuning and deployment of a parallel application in a computing infrastructure, (2) perform the tuning of the application efficiently in terms of human effort and resources needed. This paper reports the costs for carrying out the tuning of a large electromagnetic simulation based on TLM for the platform Grid'5000 and the improvements obtained on the total execution time of the application.

Keywords : Transmission-line matrix methods Load balancing Experiment methodology Large scale system Grid computing High performance computing

Type de document :

Communication dans un congrès

IEEE/ACM International Symposium on Cluster, Cloud and grid Computing ( IEEE/ACM CCGrid ), May 2014, Chicago, United States. pp.465-472, 2014, 〈10.1109/CCGrid.2014.26〉

Domaine :

Informatique [cs]

Liste complète des métadonnées

Littérature citée [20 références] Voir Masquer Télécharger

https://hal.archives-ouvertes.fr/hal-01228344
Contributeur : Thierry Monteil <>
Soumis le : jeudi 12 novembre 2015 - 22:45:31
Dernière modification le : mercredi 11 avril 2018 - 01:52:51
Document(s) archivé(s) le : vendredi 28 avril 2017 - 04:49:51

MI9_CCGRID.pdf

Fichiers produits par l'(les) auteur(s)

Platform Calibration for Load Balancing of Large Simulations: TLM Case

Fichier

Identifiants

Collections

Citation

Exporter

Métriques

496

84

Contact

Platform Calibration for Load Balancing of Large Simulations: TLM Case

Fichier

Identifiants

Collections

Citation

Exporter

Partager

Métriques

496

84

Contact