RightCapacity: SLA-driven Cross-Layer Cloud Elasticity Management
Résumé
Cloud computing paradigm has become the solution to provide good service quality and exploit economies of scale. However, the management of such elastic resources, with different Quality-of-Service (QoS) combined with on-demand self-service, is a complex issue. New challenges for elasticity management arise when people look deeper into the Cloud characteristics such as non-ignorable instance initiation time and full hour billing model. The main challenge for a SaaS provider is to determine the best trade-off between profit and end-user satisfaction. This paper proposes RightCapacity, an approach driven by Service Level Agreement (SLA) for optimizing the Cloud elasticity management (i.e., both elasticity at the application and at the infrastructure levels). We consider cross-layer (application-resource) Cloud elasticity. We model Cloud application using closed queueing network model taking into account the SLA concept and the Cloud economic model. Our results show that RightCapacity successfully keeps the best trade-off between SaaS provider profit and end-user satisfaction. Using RightCapacity, the cost saving of as much as 30% can be achieved while causing the minimum number of violations, as small as 1%.