摘要

Data integration is explored using methods of data mapping and matching using distributed heterogeneous databases. A grid environment accesses and shares the data from heterogeneous data sources or federated databases. Grid introduces service based architecture to overcome the deficiencies of traditional distributed environments related to data sharing. In response, we have elaborated the approaches, methods, frameworks, models and techniques used in existing traditional distributed environments. This helps in understanding the advantages and disadvantages of existing practices. Heterogeneous databases are selected from remote sites located in different places and data is arranged according to scenarios. Experiments are performed and scenarios are compared with different sets of data and existing methods. To measure performance and reliability of an execution experiments are also performed by selecting a large volume of data. These experiments use a toolkit for data mapping, matching and loading to facilitate both traditional and grid environments.