Personalized requirement oriented virtual service resource aggregation method

Personalization and composition are two important features of modern service eco-systems nowadays. A personalized requirements oriented virtual resource aggregation method is proposed in this paper. Service resources are formally described based on multi-dimensioned classification. Personalized requirements from applications are classified, reduced and finally expressed in formal and reusable manners. On this basis, a dynamic pruning based resource aggregation method is presented. The method pays attention to the features of both customer requirements and the organization of service resources. Multiple resources can be dynamically aggregated into coarse-grained virtual resources thus satisfy the requirements rapidly. Experiments show fairly good effectiveness and efficiency of the proposed method.