Abstract:

In order to develop an efficient integrity technology of distributed database for the evaluation of prospective cost of integrity checking,a new integrity checking strategy is proposed,and two generation algorithms respectively based on the local pretest strategy and the support test strategy are designed. Moreover,a model to evaluate the cost of integrity checking is established by using the dimension-reducing technology of artificial intelligence,which provides a base for the quantitative intercomparsion of cost evaluation of integrity checking. Then,the integrity test,the amount of accessed data,the amount of transferred data and the number of sites involved in the computation are verified by some experiments of distributed database system. The results demonstrate that the proposed strategy and algorithms effectively reduce the integrity checking cost of distributed database,and that the established cost evaluation model is of high accuracy.

Key words: distributed database, integrity test, cost evaluation, checking strategy, constraint