The Research And Implementation Of Data Integration Based On View Duplication

With the continuous development of the information technology, prompting more and morebusinesses and organizations to catch up with the trend of it. information integration technologyin order to facilitate the exchange and sharing of information. Information integration systemintegrate different geographical data of heterogeneous information sources and provides aunified interface to the users. Jinan University, High-Performance Database Project Teamproposed a new architecture of integration information system named Resource Agent Systemthat supports two ways of integration, they’re virtual integration and materialized integration.This article focuses on materialized solutions problems the integration encountered, we calledthat method is a way of collecting data based on the view data replicated. The source ofimformation supports incremental capture of the base table, and send the incremental data to theintegation side. The materialized view may be decomposed to several sub views and maintainthat.Various of factors, such as network latency, improper time of point to maintain the view,which may lead to the data in the source of information inconsistent with the data in the datawarehouse, the users and the upper applications may not tolerate this problem. this article willfocus on solving this problems. materialized maintenance, the most important step isdecomposed into sub views, decomposition algorithm introduced in detail in this article.Finally, we also do some introduction of the key module implementations in this system.