Research And Implementation Of Heterogeneous Databases Integration Framework Based On SMSQL

The integration of heterogeneous database technology has become one of the hot spots of the theory and application in the field of database. Its purpose is to structure transparent global database and shield the heterogeneity between participating databases (local database) which is geographical distributed.The paper analyses the basic principles and typical framework of the heterogeneous database integration. Based on pointing out the advantages of multi-language database system and the deficiency of traditional languages, this paper proposes the concepts of global schema, schema mapping and definition of a mulitdatabase language SMSQL (Schema Mapping SQL). The next thing is designing and implementing the SMSQL middleware (which analyze and execute the SMSQL language function) to application in specific projects.The main contents of the paper include:1. Introduce the basic concepts and objectives of the heterogeneous database integration. Provide a basis for further study by analyzing the advantages and disadvantages of the existing heterogeneous database integration technology.2. Analyze the advantages of the multi-database language system of separating the abstract and implementation, and the disadvantages of the traditional multi-databases languages (lacking of global schema, so users are hard to operate the system/ user-unfriendly). Next, the writer proposes model reuse integrated concept, defines of a mulitdatabase language SMSQ which is Schema Mapping SQL, supports the definition or modification of the global schema and also resolves the potential conflicts due to structural and semantic heterogeneity of the integrated data.3. On the basis of heterogeneous databases integration framework based on SMSQL, the writer proposes to design and implement the SMSQL middleware from several aspects including global data dictionary, SMSQL language analysis, query processing, query decomposition and so on.4. Using the SMSQL middleware in the project of grid vehicle identification application system to handle substrate heterogeneous database reduces coupling of the system, improves the development efficiency and enhances the maintainability of the system. Practical application shows that operationality, maintainability and scalability of the framework are desirable.