Enterprise Heterogeneous Database Integration Research Based On OGSA-DAI

Along with the Internet development, enterprise's practical application will possibly meet the problem that the database distributes in the different places, moreover these data is heterogeneous. Practical application need to carry on the effective integration to these distributional heterogeneous data. Simultaneously, original data had already developed application system in these data foundation accordingly. This paper developed the data integration not to be able to affect the existing system. The data integration goal is that visitors can not effect the distributional heterogeneous data integration, it does not change the original data, and the data can be issued conveniently. Moreover it guarantees that own data pool is secure. Users hope to visit the data transparently through the data integration system, need not to understand the detail of the data integration inside so long as the unified gateway can be visited.In view of the distributional heterogeneous data integration characteristic and the request, researching the globus 4.0, this paper has realized the building of grid environment in the local area network , simple CA installment and deployment, and the building of the OGSA-DAI。This paper gives the data service deployment tool, realized the data service deployment, the data resources registration, as well as data resources cancellation. The user can facilitate increase, the delete the data services .The paper has analyzed the isomerism database and database integration various roles as well as the data integration structure, realized the xml data and relational data's query unified, and designed hospital's heterogeneous database integrative system, The data come form data resource through the data service transmits data to users. Because the data service has the standard Web Service interface, this system realizes the data resources detail information hiding, achieves data visiting transparently.The paper provides reference about heterogeneous data integration under the grid environment.