QoS based Resource Allocation and Service Selection in the Cloud

Abstract

Web service composition builds a new value-added web service using existing web services. A web service may have many implementations, all of which have the same functionality, but may have different Quality of Service (QoS) values. Hence, a challenging issue of web service composition is how to meet QoS and to fulfil cloud customers' expectations and preferences in the inherently dynamic environment of the Cloud. Addressing the QoS based web service selection and resource allocation is the focus of this paper. This challenge is a multi-objective optimization problem. To tackle this complex problem, we propose a new Penalty Genetic Algorithm (PGA) to help a Cloud provider quickly determine a set of services that compose the workflow of the composite web service. The proposed approach aims to, at the one hand, meet QoS constraints prioritized by the Cloud customer and, at the other hand, respect the resource constraints of the Cloud provider. To the best of our knowledge, this is the first attempt to handle the problem of the optimal selection of web services while taking into account the resource allocation in order to guarantee the QoS imposed by the Cloud customer and to maximize the profit of the Cloud provider. The experimental results of Penalty Genetic Algorithm show that it outperforms the Integer Programming method when the number of web services and the number of resources are large.