Monte-Carlo Simulation on Heterogeneous Distributed Systems: a Computing Framework with Parallel Merging and Checkpointing Strategies - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Future Generation Computer Systems Année : 2013

Monte-Carlo Simulation on Heterogeneous Distributed Systems: a Computing Framework with Parallel Merging and Checkpointing Strategies

Résumé

This paper introduces an end-to-end framework for efficient computing and merging of Monte Carlo simulations on heterogeneous distributedsystems. Simulations are parallelized using a dynamicload- balancing approach and multiple parallel mergers. Checkpointing is used to improve reliability and to enable incremental results merging from partial results. A model is proposed to analyze the behavior of the proposed framework and help tune its parameters. Experimental results obtained on a production grid infrastructure show that the model fits the realmakes pan with a relative error of maximum 10%, that using multiple parallel mergers reduces the makes pan by 40% on average, that checkpointing enables the completion of very long simulations and that it can be used without penalizing the makespan.

Dates et versions

hal-00797962 , version 1 (07-03-2013)

Identifiants

Citer

S. Camarasu-Pop, T. Glatard, R. Ferreira da Silva, P. Gueth, D. Sarrut, et al.. Monte-Carlo Simulation on Heterogeneous Distributed Systems: a Computing Framework with Parallel Merging and Checkpointing Strategies. Future Generation Computer Systems, 2013, 3, pp.728-738. ⟨10.1016/j.future.2012.09.003⟩. ⟨hal-00797962⟩
95 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More