Queuing Theoretic and Evolutionary Deployment Optimization with Probabilistic SLAs for Service Oriented Clouds

This paper focuses on service deployment optimization in cloud computing environments. In a cloud, each service in an application is deployed as one or more service instances. Different service instances operate at different quality of service (QoS) levels. In order to satisfy given service level agreements (SLAs) as end-to-end QoS requirements of an application, the application is required to optimize its deployment configuration of service instances. $E^3/Q$ is a multiobjective genetic algorithm to solve this problem. By leveraging queuing theory, $E^3/Q$ estimates the performance of an application and allows for defining SLAs in a probabilistic manner. Simulation results demonstrate that $E^3/Q$ efficiently obtains deployment configurations that satisfy given SLAs.