Tradeoff to minimize extra-computations and stopping criterion tests for parallel iterative schemes - Archive ouverte HAL Access content directly
Reports (Research Report) Year : 2004

Tradeoff to minimize extra-computations and stopping criterion tests for parallel iterative schemes

(1, 2) , (3) , (4) , (5) , (6)
1
2
3
4
5
6

Abstract

Parallel synchronous iterative algorithms are often penalized by global synchro- nization, due to the cost of stopping tests that are achieved. It is well known that such global synchronizations are expensive for parallel implementations on distrib- uted systems, especially on clusters of processors or computational grids, where the heterogeneity and the number of processors imply a large overhead for this global operation. The aim of this work is to propose a new control technique for the stopping tests, original to the best of our knowledge, which enables to reduce the number of global synchronization near to the optimum, while keeping the number of iterations close to the number performed by standard synchronous algorithm. The main advantage of the proposed technique is that the semantic of the sequential al- gorithm is not modified, so that convergence is preserved and identical outputs are guaranteed. Our method is based on an amortized technique inspired by Floyd's and Brent's algorithms to detect periodicity in a sequence.
Fichier principal
Vignette du fichier
2004-BDMMR-tradeoffiterative.pdf (281.15 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00777293 , version 1 (29-01-2013)

Identifiers

  • HAL Id : hal-00777293 , version 1

Cite

Olivier Beaumont, El Mostafa Daoudi, Nicolas Maillard, Pierre Manneback, Jean-Louis Roch. Tradeoff to minimize extra-computations and stopping criterion tests for parallel iterative schemes. [Research Report] 2004, pp.13. ⟨hal-00777293⟩
619 View
146 Download

Share

Gmail Facebook Twitter LinkedIn More