Service Data Objects-based Heterogeneous Data Integration

With the development of internet and the gradual deepening of informationization,the demand for data sharing and data exchange constantly increase,so it is more and more pressing to integrate multiple data that generate from the process of implementing enterprise informatization to resolve the problem of "information isolated islands".Developing a heterogeneous data integration system is a valid way,however,these data sources have obvious autonomy,distribution and heterogeneity,as well as every original integration solution has its limitations,the traditional solutions have been unable to meet today's data integration needs.So in this paper a new solution of heterogeneous data integration has been proposed,which can effectively integrate multiple data sources.The main research work and content contributions of the paper are described as following:At first the theoretical foundation of data integration is reviewed.This paper also analysis and summarizes the problems which need to be resolved in the integrated process.Then it analysis and compares the traditional solutions,and points out the problems of the existing solutions.By the research of the latest programming specification—SDO(Service Data Objects) and technologies related to it,the paper proposes a solution of heterogeneous data integration based on service data objects.Then the paper introduces the data integration system architecture, and describes every sub-module's feature and implementation.Data access service(DAS) is used to encapsulate multiple data resources.It returns results by the same format,which unifies the methods of data access and data manipulation.The operation and connection of the data sources are promoted to DAS,which increases the reusability of the procedure.The problem that the traditional solutions can not write data but only read data also is resolved in this paper.It also designs and achieves a unified data access interface,which hides the differences of data sources and implements the data access in real time.Query processing is one of key issues of heterogeneous data integration system.It is directly related to the correctness and availability of data integration system.In this paper,a query decomposition method based on configuration files of data sources is given in query decomposition respect.By the configuration files,the integration system can obtain connections with the corresponding data sources.And as long as the configuration file is modified,the system can implement the flexible modification of data sources.Finally,the integrated solution is applied to the PLM system based on WEB,and its feasibility is verified.In the end of this thesis,it summarizes the work and the innovations of this paper.It also points out the future research directions and the aspects which need to be improved.