基于属性更新的语义缓存一致性维护算法

目前语义缓存一致性维护的更新策略主要是将更新操作转化成删除和添加操作，其增加了不必要的数据通信流量和数据存储，而且只能处理全属性的情况，故本文提出新的语义缓存一致性维护算法，通过分析语义缓存段与更新语句的条件谓词以及投影属性的关系，从而将更新粒度细化至被更新的属性。理论分析和仿真实验均表明该策略能够有效的减少数据存取和数据通信的开销，以及提高缓存命中率和减少查询响应时间。

Algorithm of Semantic Caching Coherency Control Scheme

In traditional semantic caching coherency control methods, update operation was usually replaced by delete operation and insert operations, but it is valid when the projection attributes are all attributes only. In order to reduce the addition of unnecessary data communications flow and data storage brought by traditional strategy, this paper developes a new algorithm for update operations based on semantic trimming. The granularity of update operations reach the exact size of the attributes by analyzing the relation between Semantic Caching segment and update operation. Theoretical analysis shows that new algorithm effectively increases the cache hit ratio and reduces data access, data communications costs and search time. Finally, The performance of this new semantic caching coherency control scheme is examined and analyzed through a simulation study, which proves it is the reliable and effective.