A fine-grained message passing MOEA/D - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2015

A fine-grained message passing MOEA/D

Résumé

We propose the first large-scale message passing distributed scheme for parallelizing the computational flow of MOEA/D, a popular decomposition-based evolutionary multi-objective optimization algorithm. We show how synchronicity and workload granularity can impact both quality and computing time, in an extremely fine-grained configuration where each individual in the MOEA/D population is mapped to a single distributed processing unit. More specifically, we deploy our distributed protocol using a large-scale environment of 128 computing cores and conduct a throughout analysis using a broad range of bi-objective combinatorial ρMNK-landscapes. Besides being able to show significant speed-ups while maintaining competitive search quality, our experimental results provide insights into the behavior of the proposed scheme in terms of quality/speed-up trade-offs; thus pushing a step towards the achievement of effective and efficient parallel decomposition-based approaches for large-scale multi-objective optimization.
Fichier principal
Vignette du fichier
cec-pmoeadxy.pdf (489.87 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-01151874 , version 1 (04-01-2017)

Identifiants

  • HAL Id : hal-01151874 , version 1

Citer

Bilel Derbel, Arnaud Liefooghe, Gauvain Marquet, El-Ghazali Talbi. A fine-grained message passing MOEA/D. IEEE Congress on Evolutionary Computation (CEC 2015), May 2015, Sendai, Japan. pp.1837-1844. ⟨hal-01151874⟩
270 Consultations
142 Téléchargements

Partager

Gmail Facebook X LinkedIn More