Always Best Served: On the behaviour of QoS- and QoE-based algorithms for Web service adaptation

Abstract

One of the primary issues in pervasive computing is the adaptation of the communication to the needs of limited devices. Web services are one of the technologies that present both big challenges and big potentials when it comes to their adapted usage in pervasive systems, because they carry communication overhead in order to support interoperability and platform-independence through self-description. In this paper it is explained how Web service communication can be adapted for limited devices and why it is important to choose intelligently among a variety of adaptation mechanisms, based on the system context. Further, two algorithms (one Quality of Service-based and one Quality of Experience-based) for the decision support of the mentioned problem are provided in order to measure and discuss the impact that the peculiarities of the problem have on their behaviour.