Toward Informed Resource Management in the Cloud

Giang Son Tran (1) , Laurent Broto (1) , Daniel Hagimont (1, 2)

1 IRIT-SEPIA - Système d’exploitation, systèmes répartis, de l’intergiciel à l’architecture
2 Toulouse INP - Institut National Polytechnique (Toulouse)

More and more companies are externalizing their computing infrastructures to the cloud to reduce the increasing maintenance cost of computing environments. Minimizing the amount of hardware resource and power consumption in use is one of the main services that such a cloud infrastructure must ensure. This objective can be done either by the customer at the application level (by dynamically sizing the application based on the workload), or by the provider at the virtualization level (by consolidating virtual machines based on the infrastructure's utilization rate). Many research works investigate resource management policies separately. In this paper, we show that the different strategies for cloud resource management, including server consolidation only, dynamic application sizing only, both policies at the same time, do not fully bring benefits to the cloud actors when being implemented without cooperation. Finally, we propose and evaluate a cooperative model to combine the efficiency from these strategies in reducing power consumption and keeping application's Quality of Service.

Mots clés: Cloud resource management; Cloud

Domaines: Réseaux et télécommunications [cs.NI]; Système d'exploitation [cs.OS]; Architectures Matérielles [cs.AR]; Systèmes embarqués

Giang Son Tran, Laurent Broto, Daniel Hagimont. Toward Informed Resource Management in the Cloud. IEEE/ACM International Conference on Utility and Cloud Computing - UCC 2013, Dec 2013, Dresden, Germany. pp. 247-250. ⟨hal-01151022⟩