针对当前服务组合研究主要考虑单个用户请求而忽视并发请求的问题,提出一种基于QoS模糊支配的并发服务组合方法。根据QoS模糊支配关系,提出模糊支配得分的概念,以衡量服务的整体质量。在服务组合过程中,从每个组件中选出模糊支配得分最高的k个服务,去匹配批量的用户请求。对于每个组合请求,从每个相关组件的top-k服务中随机选择一个服务参与服务组合。实验结果表明,该方法可为用户提供较高质量的组合服务,具有较好的实时性。
Aiming at the problem that most existing service composition approaches focus on the single user's request and ignore the concurrent requests, a concurrent service composition approach based on the QoS fuzzy dominance was proposed. According to the strength of the QoS fuzzy dominance relationship, a concept called fuzzy dominance score was proposed to measure the overall quality of a service. In the service composition process, k services with optimal fuzzy dominance score were selected from each component to perform bulk users~ requests. A service was selected randomly from the top-k services for each involved com- ponent to take part in service composition for every request. Experimental results show that the proposed approach can provide users with combined services of high quality and has good real-time performance as well.