SchemaSQL - A Language for Interoperability in Relational Multi-Database Systems
Laks V. S. Lakshmanan, Fereidoon Sadri, Iyer N. Subramanian · 1996
We provide a principled extension of SQL, called SchemaSQL, that offers the capability of uni-form manipulation of data and meta-data in re-lational multi-database systems. We develop a precise syntax and semantics of SchemaSQL in a manner that extends traditional SQL syntax and semantics, and demonstrate the following. (1) SchemaSQL retains the flavour of SQL while sup-porting querying of both data and meta-data. (2) It can be used to represent data in a database in a structure substantially different from origi-nal database, in which data and meta-data may be interchanged. (3) It also permits the cre-ation of views whose schema is dynamically de-pendent on the contents of the input instance. (4) While aggregation in SQL is restricted to values occurring in one column at a time, SchemaSQL permits “horizontal ” aggregation and even aggre-gation over more general “blocks ” of informa-tion. (5) SchemaSQL provides a great facility for interoperability and data/meta-data manage-ment in relational multi-database systems. We provide many examples to illustrate our claims. We outline an architecture for the implementa-tion of SchemaSQL and discuss implementation algorithms based on available database technology that allows for powerful integration of SQL based relational DBMS.