A decentralized convergence detection algorithm for asynchronous parallel iterative algorithms - Archive ouverte HAL Access content directly
Journal Articles IEEE Transactions on Parallel and Distributed Systems Year : 2005

A decentralized convergence detection algorithm for asynchronous parallel iterative algorithms

Abstract

We introduce a theoretical algorithm and its practical version to perform decentralized detection of the global convergence of parallel asynchronous iterative algorithms. We prove that even if the algorithm is completely decentralized, the detection of global convergence is achieved on one processor under the classical conditions. The proposed algorithm is very useful in the context of grid computing in which the processors are distributed and in which detecting the convergence on a master processor may be penalizing or even impossible as in Peer to Peer computations framework. Finally, the efficiency of the practical algorithm is illustrated in a typical experiment.
Fichier principal
Vignette du fichier
ConvDev_hal.pdf (193.91 Ko) Télécharger le fichier
Loading...

Dates and versions

hal-00096362 , version 1 (19-09-2006)

Identifiers

  • HAL Id : hal-00096362 , version 1

Cite

Jacques M. Bahi, Sylvain Contassot-Vivier, Raphaël Couturier, Flavien Vernier. A decentralized convergence detection algorithm for asynchronous parallel iterative algorithms. IEEE Transactions on Parallel and Distributed Systems, 2005, 16, pp.4--13. ⟨hal-00096362⟩
70 View
279 Download

Share

Gmail Facebook Twitter LinkedIn More