Publish/Subscribe Driven Matching Algorithm

The characteristics of publish/subscribe technology are of asynchronism and loosely coupled. This phenomenon makes producers and consumers are de-communicated in temporal and spatial domains during processing. It requires develop a dynamic integration in today’s large-scale, distributed information system. This paper introduces a Semantic Integration Data Model (SIDM) for publish/subscribe systems, within which model subscription module is designed. The model contains a matching algorithm which combines both publishing-driven and subscription-driven matching algorithms. Our experiments show that this matching algorithm has high efficiency for matching.