version française rss feed
HAL : hal-00757074, version 1

Voir la fiche détaillée  BibTeX,EndNote,...
Proceedings of the 21st ACM International Symposium on High Performance Distributed Computing (HPDC'12), Delft : Pays-Bas (2012)
SpeQuloS: A QoS Service for BoT Applications Using Best Effort Distributed Computing Infrastructures
Simon Delamare1, Gilles Fedak1, Derrick Kondo2, Oleg Lodygensky3

Exploitation of Best Effort Distributed Computing Infras- tructures (BE-DCIs) allow operators to maximize the uti- lization of the infrastructures, and users to access the un- used resources at relatively low cost. Because providers do not guarantee that the computing resources remain available to the user during the entire execution of their applications, they offer a diminished Quality of Service (QoS) compared to traditional infrastructures. Pro ling the execution of Bag-of-Tasks (BoT) applications on several kinds of BE-DCIs demonstrates that their task completion rate drops near the end of the execution. In this paper, we present the SpeQuloS service which en- hances the QoS of BoT applications executed on BE-DCIs by reducing the execution time, improving its stability, and reporting to users a predicted completion time. SpeQuloS monitors the execution of the BoT on the BE-DCIs, and dynamically supplies fast and reliable Cloud resources when the critical part of the BoT is executed. We present the design and development of the framework and several strategies to decide when and how Cloud resources should be provisioned. Performance evaluation using simulations shows that SpeQuloS ful ll its objectives. It speeds-up the execution of BoTs, in the best cases by a factor greater than 2, while offloading less than 2.5% of the workload to the Cloud. We report on preliminary results after a complex deployment as part of the European Desktop Grid Infrastructure.
1 :  LIP Lyon / Inria Grenoble Rhône-Alpes - AVALON
2 :  INRIA Grenoble Rhône-Alpes / LIG laboratoire d'Informatique de Grenoble - MESCAL
3 :  LAL - Laboratoire de l'Accélérateur Linéaire
Informatique/Calcul parallèle, distribué et partagé