Research On Heterogeneous Data Exchange Model Based On XML

With the improvement of information level, government agencies and business units gradually established their own information management system to store and process a large amount of data. The information management system based on different network environment and database systems, makes abundant data with different formats, structure and semantics. Thus the heterogeneity is formed between multiple data source.Data exchange is produced to carry out effective data flow and information sharing between heterogeneous data sources. EDI, data warehouse method and other mainstream heterogeneous data exchange technology is limited by low coupling and expansion. Though the traditional exchange model based on xml has overcome these shortcomings, it breaks the data integrity because user-defined constraints is not noted completely. This subject proposes a new heterogeneous data exchange model based on XML. Not only including the advantages of the traditional exchange model, expanding the data structure constraints, summarizing the type of check constraint, adding the self-growth constraints, and thus it enhances the constraints integrity of the exchange model.There are several aspects of this research.Firstly, comparing with several major data exchange technology, this subject introduce the characteristics and related technology of XML, and promote the traditional exchange model.Secondly, the design of exchange model in this subject includes network topology and architecture design, frame construction and module division.Thirdly, the data exchange contains relational data, that is data records in the table; and the relational schema including data structures and constraints. NOT only constructing models for relational data and relational schema, establishing mapping relationship, improving the traditional exchange algorithm, this model achieve exchange between multiple data source through the conversion between data source and XMLFourthly, when the number of table data records is big enough, the transformed XML document will be big too. That makes network congestion in the data transfer. The data partitioning method designed in this subject can partition XML documents from the view of record and schema, optimizing network status.This exchange model can eliminate data heterogeneity. The experiment between SQL Server 2008 and Oracle 11g confirms the integrity and accuracy of the data exchange. It provides theoretical basis and support technique for effective data exchange in government agencies and business units.