Research On The Model For Web Service Composition Based On QoS And Providers' Benefit

Web Services technology is based on some technical standards such as XML, SOAP, WSDL and UDDI. It is a self-contained, self-describing, modular technology. It realizes the platform independence and the programming language independence, thus it solves some technical problems such as application integration and data sharing in heterogeneous platforms. With the rapid development of Web Services technology, this technology has been widely used. The users of Web Services require better QoS and the providers of Web Services expect more benefit by providing Web services.The development of Semantic Web Services technology provides a technical basis for QoS-based Web Services discovery, selection and composition. However in order to ensure QoS and service providers'benefit in SOA applications in business domains, there are some key technologies need to be studied. Under normal circumstances the information of QoS is provided by Web services provider, thus it is difficult to guarantee the objectivity of the QoS information. In existing QoS management studies, the final QoS value is computed to reflect the general QoS of a Web service. Such a single final QoS value cannot reflect QoS trend. Usually there are some inaccurate feedback evaluation data of QoS cased by subjective or objective factors. How to guarantee QoS and maximize the providers'benefit is also a problem.To deal with the problems above-mentioned a model for Web Services composition based on QoS and providers'benefit is studied. An algorithm for QoS trend evaluation and an algorithm to eliminate the errors in QoS feedback evaluation data are realized in this model. The two algorithms are adopted to enhance the veracity of QoS evaluation. This model realizes QoS-based Web Services discovery, selection and composition, guarantees the availability of Web Services composition. The Web Services request scheduling algorithm in this model maximizes service providers'benefit.