基于树的解析几何多维事件匹配算法

In a large-scale content-based information distribution system, event matching is the core component, which is responsible for finding all subscriptions that match an event. However, as the number of users continues to increase and messages become richer and more diverse, most existing event matching algorithms cannot effectively subtain system performance.This paper presents GEM-Tree(Geometrical Event Matching Tree), a tree-based analytic geometric event matching algorithm that can be used for efficient event matching in large-scale content-based information distribution systems.To further improve the event matching speed, a local-adjustment mechanism is designed to determine the deployment for each new subscription registering into the GEM-Tree, and a global-adjustment mechanism is designed to optimize the location of the subscriptions already inserted in GEM-Tree.In the various scenarios of the experiment, GEM-Tree is superior other algorithms (BE-Tree, OP-Index, and TAMA).