QoS-Driven Service Selection

Web service is a new application model for decentralized computing, and it is also an effective mechanism for the data and service integration on the web. The complex web service is made of the web service components and the element web services. Web service composition offers a simple way for the creation of complex applications.The current research of web service composition focuses on the automated composition. It includes automatically generating service execution plan and selecting services. There are so many same services in the Internet that one of them has to be selected to run in the execution plan. The semantic technology is importance of automated generation of the composition plan and service selection.The concrete service which executes in runtime is decided by the non-function properties of services. The services which have the same semantic information have different non-function properties. And the composition of services is under the user or service designer constraints. The concrete execution plan is able to be realized by the Quality of Service (QoS).The service would fail for various factors when a composition service runs in the dynamic environment. So another service which has the same semantic information to the failed service should be invoked. Re-planning such a composition may be needed whenever deviations from the QoS estimates occur. It is necessary for the composition's QoS and selection algorithm to follow the constraints. These are discussed in this paper.The Selection Algorithm of Assuring QoS (SAAQ) the paper proposes is an approach for QoS-aware service composition that bases genetic algorithms. It selects services with their QoS. It estimates the execution plan by fitness function. It searches the potential solution according to the crossover operator, the mutation operator and the selection operator of genetic algorithms.The paper works at the overall QoS of a composition by aggregating the QoS of the individual services. And the algorithm maps the values of QoS into a constant length closed interval for estimating individual fairly, because the values of QoS are various. SAAQ inserts the best solution to a queue from the population at each evolution step. And then the queue of...