Research On Heterogeneous Data Integration Technology Based On XML

with the development of information construction and network popularization, a considerable number of enterprises have produced more and more heterogeneous data storied in different platforms, depended on different database management system(such as XML document, text document, etc). How to integrate and use these heterogeneous data with different format from different sources through the network is a needed way in the process of enterprise development. Only in this way can it share the data completely in enterprise in order to make it play a greater role. So, it is necessary to do some research on heterogeneous data integration to meet this demand.Firstly, this paper focuses on theories and technologies of existing heterogeneous data integration method, the XML technology and the Web Services technology. Then, a kind of heterogeneous data integration architecture based on XML by using Mediator/ Wrapper approach is designed.This structure adopts XML Schema to build public models, uses XML as the exchange data format, uses XQuery as the global query language, masks the platform discrepancy of data sources environment based on Web Services technology, to set up a public integrated environment in the mediation which shields all differences in aspects of data source including the platform, system environment and internal data structure, and so on. It provides a unified and transparent interface for user and implemented the accession of integrated query.Secondly, this paper designs and describes the function of each module in the system, and defines the key modules. Then, the method for transforming from the relation schema to XML Schema model is designed and implemented; a public model of the integration system is constructed. An algorithm for global decomposed querying based on the public model in the query processor has been proposed. It makes the system less coupling, flexible and scalability through encapsulating each heterogeneous data source to a web service. Finally, a student information querying system has been implemented by using this framework, which can complete the heterogeneous data query function including structured and Semi-structured data sources. The experiment result shows that the architecture designed in this paper is feasible and validated.