Research On Integrity Constraints In Integrated Data

Information integration aims to provide a unified global view for users to access multiple heterogeneous information sources, to shield data sources, to overcome the heterogeneity and data conflicts, so as to provide transparent access to the application of the data. The so-called heterogeneity includes the data model, data management system technology differences and logical heterogeneity, which includes the logic of heterogeneous model and semantic differences.Integrity constraints provide a way to maintain semantic consistency between data in database and external reality. In the traditional database(including the ideal distributed database), integrity constraints guarantee that the user will not destroy the consistency of the data when the database is modified by the authorized user. That when an application attempts to contain insert, delete and update statements update transaction changes the state of the database, the DBMS will to possible new state in accordance with the integrity constraints for a given inspection, and to resist that led to the integrity constraints is the destruction of the events. In order to ensure the effectiveness of the data obtained by the integrated system, it is necessary to impose the necessary integrity constraints on the integrated database described in the global schema. Due to the integrated database is only relevant information source data integration and fusion, information source system of autonomous and large-scale data integration in system complexity are not allowed to integrated database integrity maintenance as the ideal distributed database that through the information source update affairs global integrity inspection to achieve; at the same time, this also is not necessary. In principle, it is necessary to ensure the integrity of the integrated database to ensure that the results obtained from the query are satisfied with these constraints.In the past in data warehouse(essentially is a kind of entity integrated) background to the development of the topic of data quality assurance data cleaning, and later for the virtual integration development of consistent query can be regarded as service to this goal in different integration method take the means of realizing the. The problem is that all of these techniques are based on the constraints provided by the query itself to test and repair.(note that entity integration can be seen as an entity store for query results). This integration of virtual integration and virtual and physical integration is far from solving the problem. The first is how to generate constraints on the query? The second is how to distribute the task of integrity test and data repair to the components of the information integration system.To solve these problems, this paper includes the following two aspects:(1) to improve the integrated system of global end integrity constraint processing efficiency, the complete constraint maintenance tasks between the integrated terminal and each information source adapter for reasonable load distribution is global complete Constraint Decomposition to maintain the integrity of the local mode and terminal integrated data conflict resolution. In this paper, based on the idea of Constraint Decomposition, this paper discusses the problem of constraint propagation from global mode to local mode.(2) in order to solve the data inconsistency in the process of integration, a new algorithm is proposed. Based on the heuristic algorithm, the global end of the data is not consistent with the data from different local patterns.(3) through the deployment of the implementation of the integrated system, the implementation process of the integrity constraints is verified, the Constraint Decomposition and repair algorithm proposed in this paper is implemented and applied.