An Intelligent Admission Control Framework for Composite Services

Web service composition is a new technology for creating value-added services in the dynamic network environment and implies great application value, with the rapid development of Web services and Cloud Computing, composite service requests are growing sharply. However, how service integrators intelligently to handle continuous arrival of composite service requests, ensure service quality of existing users while improving resource utilization is a challenge for urgent solution. This paper provides an intelligent admission control framework for composite services, which mainly including three aspects: Firstly, it proposes a global QoS constraints optimal decomposition method which decomposes global QoS constraints into local QoS constraints that each task in the composition process should satisfy, this step provides critical basis for admission judgment, Secondly, it proposes a reliable admission judgment method based on QoS dynamic prediction and QoS interference estimation, this method can improve the reliability of admission control. Finally, it provides a context-aware service resource adaptive scheduling method for assigning the most suitable resource to the new accepted request so as to improve the utilization of the whole service resources. Our study aims to reveal the general rules of admission control for composite service, provide methods to improve the quality of composite service and increase the utilization of the whole service resource.