Research on QoS-based service selection

Abstract

Web services as a key technique for implementing service-oriented architecture (SOA) has been attached with more and more importance. With the rapid development of Web services, quite a few services are emerging, which provide the same functionality but may differ in QoS. Therefore, subsequent works have focused on how to select the desired services from all functionality-similar services. Considering service users’ preferences, QoS constraints, and the dynamics of QoS attributes in the Web service world, service selection is really a big challenge. In this dissertation, we extend service selection to different phases of service implementation, which takes into consideration replication-based fault tolerant service selection, top-k service selection and reselection of service while satisfying QoS constraints. The main research works and contributions of this dissertation are as follows: 1. In the dynamic Web service environments, the performance of the Internet is unpredictable; the reliability and effectiveness of remoteWeb services are also unclear. Therefore, it can hardly be guaranteed that the QoS attributes of Web services do not fluctuate with the dynamic Web service environments. When a composite service is planned in the context of dynamic service environments, there is another aspect which has not been taken into account by existing works, namely, concurrency - the fact that multiple requests to a composite service may arrive at the same time. Therefore, we present a service selection scheme, which adopts top-k dominating queries to generate a composition solution rather than only select the best composition solution for a given request. Since we take into account the dynamics of Web service environments and concurrency of requests in this service selection scheme, we can avoid the situation, in which the current-best service with regard to QoS attributes may degrade a lot due to lots of requests designated to it. The experimental results have investigated efficiency and effectiveness of our approach and shown that it outperforms baseline and traditional methods for service selection. 2. Service reselection needs to be made when the current service (usually composite service) fails. In order to avoid re-constructing the whole business process, most works focus on just replacing the failed atomic service contained by the business process. However, it may not solve the problem, due to constraints violation or unavailability of remote services (server crash down, network disconnect, and so on). We propose an approach for service reselection based on constraints partition in web service composition. Our service reselection is to replace a failed service and (several) related services as a whole in the composition by other alternative services. Specifically, the approach is divided into three steps, namely, decomposing the global constraints, filtering the candidates and replacing the failed service. We evaluate our approach by comparing it with other approaches. The results of experiments show that our approach outperforms the others, especially when the number of candidate services is large. 3. We design a detecting model to judge whether the chosen services are independent of each other in the replication-based fault tolerant service selection. This model introduces a component TTP (Trusted Third Party), which adopts homomorphism cryptosystem to keep the privacy of services to be detected, so as not to disclose their inner implementation details. Specifically, we model our detecting work as a special Set-Intersection problem. By solving it, we can achieve replication-based fault tolerant service selection by using reliable web services without violating their autonomy and privacy at the same time.

Date of Award	3 Oct 2012
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Qing LI (Supervisor)