Modeling and Simulation of QoS-aware Web Service Selection for Provider Profit Maximization.

Web Services (WSs) are a new breed of web application that is gaining momentum. One of the research challenges it brings along is the establishment of an interoperable semantic framework suitable to represent not only the functional, but also the non-functional features of WSs, i.e. their Quality-of-Service (QoS), which in most cases are not included in the WS description. Nevertheless, integrating QoS features in WS profiles is to the advantage of both users and providers, as it supports QoS-aware WS selection and composition addressing the user's QoS requirements, while enabling WS providers to increase their profits. The purpose of this paper is twofold. On the one hand, it aims to present a QoS ontology framework adequate for WS provision. On the other hand, it studies the problem of providers that receive concurrent requests of numerous customers for WSs demonstrating different bandwidth and price properties. It introduces the "Selective Multiple Choice Knapsack Problem" that aims to identify the services that should be delivered to the customers in order to maximize the provider's profit, subject to maximum bandwidth constraints on the server side. Thus, the QoS-awareness in this version of this problem is limited to the consideration of two QoS parameters: service bandwidth and cost. The problem is solved by a proposed algorithm that has been empirically evaluated via numerous experiments. [ABSTRACT FROM AUTHOR]