Computer Science · Computer Networks and Communications
Multidatabase and Interoperable Systems
See on the mapAbstract
Most large organizations have many different kinds of hardware and software. There are various large and small databases scattered throughout the network with different data models and access languages. Each user has a local data source or two that is strategic to their function. However, they also realize that there is more information out in the network, somewhere, that they need. These organizations need a data access system that can accommodate preexisting, heterogeneous, data sources scattered through the network. Multidatabase systems provide a uniform global database interface to exactly these types of systems. This global interface hides all the underlying distribution and heterogeneity of information in the system, so that users can submit a single query to capture data from any or all available databases. In addition, multidatabase systems allow participating local databases to remain autonomous; i.e., the local DBMS determines how to structure local data, what functions to expose to the multidatabase, what global queries it will process, and when it will join or leave the global system. This property of local autonomy means the multidatabase system provides global information access, but the organization does not have to replace its hardware, software, or user training. Organizations keep all their existing function, while adding significant global access capability.