Mining Attribute Association in Query Predicates for Access Path Generation
Proceedings of the 2012 ACM Research in Applied Computation Symposium
ISSN or ISBN
This paper presents a framework to mine query predicates from query patterns in order to produce candidate attributes from query predicates for index generation and access path selection. Query predicates are query conditions or constraints involving tables to be accessed and values of attributes or columns to be retrieved in the format of the tuples or objects. Access path and its auxiliary data structures determine how data tuples and objects to be retrieved efficiently, in turn, the performance of data query processing. This framework mines set of query patterns, analyzes the association of attributes, and provides heuristics for access path generation and selection.
Sun, Junping, "Mining Attribute Association in Query Predicates for Access Path Generation" (2012). CEC Faculty Articles. 507.