Heterogeneity Considered Harmful to Algorithm Designers

Olivier Beaumont 1 Vincent Boudet 1 Arnaud Legrand 1 Fabrice Rastello 1 Yves Robert 1
1 REMAP - Regularity and massive parallel computing
Inria Grenoble - Rhône-Alpes, LIP - Laboratoire de l'Informatique du Parallélisme
Abstract : In this paper, we deal with algorithmic issues on heterogeneous platforms. We show that static scheduling and load-balancing strategies are absolutely needed to achieve good performances, in contrast to situation for homogeneous parallel machines where dynamic schemes often turn out to be very satisfactory. However, we also show that static strategies targeted to heterogeneous platforms are difficult to design and implement: intuitively, data distribution must obey a much more refined model than standard block-cyclic distributions to equally balance the load between processors of different speeds. Technically, we state several NP-completeness results that demonstrate the intrinsic difficulty of static load-balancing on heterogeneous platforms.
Type de document :
Communication dans un congrès
Cluster'2000, 2000, Unknown, IEEE Computer Society Press, pp.403―404, 2000
Liste complète des métadonnées

Contributeur : Arnaud Legrand <>
Soumis le : lundi 18 février 2013 - 11:52:06
Dernière modification le : vendredi 23 novembre 2018 - 13:40:03


  • HAL Id : hal-00789464, version 1



Olivier Beaumont, Vincent Boudet, Arnaud Legrand, Fabrice Rastello, Yves Robert. Heterogeneity Considered Harmful to Algorithm Designers. Cluster'2000, 2000, Unknown, IEEE Computer Society Press, pp.403―404, 2000. 〈hal-00789464〉



Consultations de la notice