The Design And Implementation Of The Management And Maintenance Tools Of Metadata Databases For Data Integration

In the 21st Century, life sciences have been developed at a tremendous speed, and biological information is expanding violently. One of the most difficult problems biologists have long faced is to find an effect method of integrating and querying the databases which are distributing, heterogeneous, and autonomic. In order to solve the problem in integrating and querying of biological information, we proposed the method of integrating protein data sources based on the metadata, and established the metabase, which included the source database structural metadata according to the CWM criterion, the user schema for the user's query and the semantic metadata according to the label of ontology. Above that, this thesis pays most attention to how to manage and maintain the metadata depository, how to synchronize the metadata between the source database, how to be sure the updating coherence between the metadata depository and the metadata and so on.Because our integrating project of the data sources is based on the data integration of metadata, we encountered a tough problem of how to effectively manage and maintain the metadata integrated into the metabase. Firstly, we must find a method to manage and maintain the metadata integrated into metabase from various databases, including the browsing, querying, and backing up of the metadata, Secondly, we must find an approach to capture and renew the structural changes into the metabase since the manager of the metabase is unable to control the changes of the source database, which is the thesis's conundrum. Based on the points raised before, the content of this thesis can be generalized as follows:1) Metadata and CWM were researched and analyzed. We used a uniting interface to implement the functions of initializing the metabase and of browsing and querying of the metadata, also realized the initialization of the ontology information and user schema information, and achieved the multi-angle browsing and querying of the user schema information and ontology information.2) Through the research of database information synchronization, we analyzed and designed the capture tools of the source database structure, updated the metabase according to the captured structure changes of the source database, renewed the user schema depository and ontology depository information according to the changes, and successfully solved the suspension problem of user schema and ontology data caused by metadata update.3) We analyzed the disaster-tolerant strategy and provided a disaster-tolerance mechanism to the metabase.Integrated the above-mentioned research, finally we designed and implemented a metadata depository management and maintenance system (CWMMS), so the metabase manager updates and maintains the metabase conveniently. Meanwhile, my work is the important part of the integrating project of the protein data which based on metadata that prepares for the farther research.