QoS-based Web Services Selection

To adapt to the dynamic market environment which is increasingly competitive, more and more corporations start to build their business flows on the Internet, and publish their own business as Web Services to customers. However, how can Services providers promote customers satisfaction when a great quantity of Services meet consumers'needs becomes the present research hot topic.In practice, consumers often do not want to spend a lot of time to find the best Web Services, but hope to quickly get the Web Services which they are satisfactory. QoS is becoming standard of Web Services differentiating and evaluating the candidate Web Services when they suit the functional requirements of consumers. In this paper, a novel model to improve Web Services selection efficiency for QoS is proposed through research Web Services selection technology and based on analysis of existing research results and paredo principle,which can improve Web Services selection efficiency .This paper has the following contribution:(1)First of all, the concept of Web Services and QoS and objective of Web Services selection for QoS are presented; and then the impotance of Web Services selection for QoS is introduced; and the detail current research situation on Web Services and QoS and Web Services selection for QoS are analyzed and summarized.(2)The current Web Services selection is only based on functionality, without taking QoS selection into account. To solve these problems, main function and basic operation of various components consisted of Web Services system are introduced.and Web Services architecture and core technologies are discussed.(3)As Fuzzy data is hard to quantify in the Web Services, fuzzy comprehensive evaluation method is used to quantify. The virtual qualitative variables related to membership functions, and weights and aggregation formula are calculated to do single factor fuzzy comprehensive evaluation and multi-level fuzzy comprehensive evaluation when Web Services QoS is in a comprehensive evaluation. Finally a calculation of fuzzy QoS attribute is illustrated by an example.(4)As a crucial technology in Web Services selection, QoS plays a key role in Web Services selection. For Web Services QoS issues, QoS attributes are introduced, different consumers can choose needed QoS attribute information from different angles. QoS Qualitative attribute is quantified by fuzzy comprehensive evaluation. And then a computational model of QoS is presented, including the standardization and the weighted average of QoS of type of description and district. Finally QoS calculation is illustrated by a example.(5)Because the current WSDL and UDDI could not support description of QoS attributes of Web Services, the Web Services WSDL and UDDI specifications are extended to support QoS mechanism. A novel Web Services selection model is designed to solve the problem of Web Services selection by improving existed models for Web Services discovery and paredo principle. To prevent duplicate calculations of Web Services selection and improve Web Services selection efficiency, the new interactive time series and Web Services selection steps are presented. Simulation experiments indicated that the proposed methods could quickly select most Web Services and effectively improve Web Services selection efficiency. In the end, the conclusion is made and the future research is put forward.