As for the rapid extraction of source data, the rapid identification of changing data and the rapid incremental extraction of data, underlying the analysis of the working principle of the traditional shadow table, this paper proposed a kind of improved linear algorithm which merged MD5 algorithm into the ST. It scanned compared tables for the linear and eliminates unnecessary inverse operation. During scanning these tables, it employed MD5 algorithm to calculate the whole record ' fingerprint ' , it reduced the frequency and duration of string matching, and it could quickly recognize the changed records. The algorithm was verified practically, and the result shows that the proposed algorithm can improve the efficiency of data extraction in database. Incremental extraction method based on shadow table is a general incremental capturing method, and it can be implemented in any database. These applications can easily be transplanted in all kinds of platforms, so it is suitable to solve the heterogeneous database replication issues.%为了快速提取源头数据、快速识别变化记录以及实现数据的快速增量提取,在剖析传统影子表法的工作原理上,提出基于MD5算法的影子表法的改进型线性算法,对对比表进行线性扫描,排除了不必要的回扫操作；同时通过MD5算法计算整条记录的“指纹”,降低了字符串比对次数和时间,能够迅速识别出发生变化的记录.对所提出算法进行了应用测试,结果表明通过融入MD5算法后的影子表法提高了数据提取效率.基于影子表的增量提取方法是一种通用的增量捕获方法,能在任何数据库上实现；应用程序可以方便地在多种平台间移植,因此很适合解决异构数据库复制问题.
展开▼
