QoS-Based Web Service Selection for Multiple Users by Genetic Algorithm

摘要

For ensuring the efficient operation of a service-oriented application, it is necessary to resolve the problem of QoS-based web service selection for multiple users. Its purpose is to choose an optimal solution, where each task forming an application's workflow is performed by one of its candidate services which have the best quality of service (QoS), for each user depending on all users' preferences for the application. Although many approaches have been proposed to address this problem, they do not incorporate different users' preferences or undertake too large overhead. For the sake of raising the effectiveness of QoS-based service selection for multiple users, we uses the Advanced A-Fully Polynomial Time Approximation Scheme (A²- FPTAS) to compute the approximate Pareto set which contains the Pareto optimal solutions of the workflow. In addition, we adopt genetic algorithm to choose the best solutions from the Pareto set for all the users in this paper. The experiment is conducted to show the validity of our approach.