Index Tuning is concerned with when and how to construct an index.
When relational models were first introduced, customers complained that vendors had introduced them to slowed down customer applications in order to make the customers buy more hardware. Vendors and researchers responded to the performance challenge by incorporating existing data structures (notably B-trees and hash structures) and improving many new ones (notably bit vectors and multi-dimensional indexes). The decision of which data structures to create was left to the user. This entry concerns that decision.
An index is a data structure plus a method of arranging the data tuples in the table (or other kind of collection object) being indexed. The syntax for constructing indexes is discussed elsewhere. This entry presents the tuning considerations.
Two data structures are most often used in practice for indexes: B+-trees and Hash structures. Of these, B+-trees are used the...
- 1.Celko J. Joe Celko’s SQL for Smarties: Advanced SQL Programming (3rd edn.). Morgan Kaufmann, San Fransisco, CA, 2005.Google Scholar
- 2.Kimball R. and Ross M. The Data Warehouse Toolkit: The Complete Guide to Dimensional Modeling (2nd edn.). Wiley, New York, NY, 2002.Google Scholar
- 3.Shasha D. and Bonnet P. Database Tuning: Principles, Experiments and Troubleshooting Techniques. Morgan Kaufmann, San Fransisco, CA, 2002.Google Scholar
- 4.Tow D. SQL Tuning. OReilly, Sebastopol, CA, 2003.Google Scholar