Schema Evolution

Since change is a fundamental aspect of persistent information and data-centric systems, both data and schema tend to change over time for a multitude of reasons, including to reflect a change in the real world, a change in the user’s requirements, mistakes in the initial design or to allow the expansion of the application scope over time. Schema evolution is a technique aimed at maintaining the extant data after applying schema changes, which has been investigated widely both in the context of static and temporal databases. While schema changes are inevitable during the life of a database, most of the current database management systems (DBMS) unfortunately do not provide enough support for these changes and do not support schema evolution. Hence, database developers and designers try to solve the problem of schema evolution in an ad hoc manner. Note that schema evolution is partially supported by some relational DBMSs, such as Oracle and Microsoft SQL Server, and also by some object-oriented DBMSs, such as Orion and TIGUKAT. With the growing interest in XML and XML data as well as the mechanisms for holding such data, the XML context within which data items are formatted also becomes an issue. Much research work has recently done on the problems of schema evolution in XML databases. The purpose of this chapter is (i) to present the different research proposals that deal with schema evolution, and (ii) to discuss the support of existing DBMSs to handle schema changes.