Current methods of evaluating nested queries in the SQL language can be inefficient in a variety of query and data base contexts. Previous research in the area of nested query optimization which sought methods of reducing evaluation costs is summarized, including a classification scheme for nested queries, algorithms designed to transform each type of query to a logically equivalent form which may then be evaluated more efficiently, and a description of a major bug in one of these algorithms. Further examination reveals another bug in the same algorithm. Solutions to these bugs are proposed and incorporated into a new transformation algorithm, and extensions are proposed which will allow the transformation algorithms to handle a larger class of predicates. A recursive algorithm for processing a general nested query is presented and the action of this algorithm is demonstrated. This algorithm can be used to transform any nested query.
[1]
Patricia G. Selinger,et al.
Access path selection in a relational database management system
,
1979,
SIGMOD '79.
[2]
Donald D. Chamberlin,et al.
Implementation of a structured English query language
,
1975,
CACM.
[3]
Irving L. Traiger,et al.
System R: relational approach to database management
,
1976,
TODS.
[4]
Won Kim,et al.
On optimizing an SQL-like nested query
,
1982,
TODS.
[5]
E. F. Codd,et al.
Extending the database relational model to capture more meaning
,
1979,
ACM Trans. Database Syst..
[6]
Michael Stonebraker,et al.
The design and implementation of INGRES
,
1976,
TODS.