Towards a novel smart and energy-aware service-oriented manager for extreme-scale applications

  • Authors:
  • Mohammed el Mehdi Diouri;Olivier Gluck;Laurent Lefevre

  • Affiliations:
  • Laboratoire de l'Informatique du Parallélisme, UMR 5668 CNRS, ENS Lyon, INRIA, Université de Lyon, 46, Allée d'Italie, 69364 Cedex 7, FRANCE;Laboratoire de l'Informatique du Parallélisme, UMR 5668 CNRS, ENS Lyon, INRIA, Université de Lyon, 46, Allée d'Italie, 69364 Cedex 7, FRANCE;Laboratoire de l'Informatique du Parallélisme, UMR 5668 CNRS, ENS Lyon, INRIA, Université de Lyon, 46, Allée d'Italie, 69364 Cedex 7, FRANCE

  • Venue:
  • IGCC '12 Proceedings of the 2012 International Green Computing Conference (IGCC)
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Exascale supercomputers will gather hundreds of million cores. The main problem to take care for running applications on such platforms is energy consumption since it is one major limitation if we consider that the currently fastest supercomputer consumes more than 12MW for a maximum performance of 10PFlops. Besides, we also need to overcome important challenges related to fault tolerance and data management in such extreme-scale systems. Thus, we need to take into consideration these challenges from an energy consumption point of view and to propose them as energy-aware services for exascale applications. At this end, we propose in this paper to provide accurate estimations of the energy consumption due to these services and offer some green and energy efficient solutions, leading to a smart and energy-aware service-oriented manager for exascale applications.