Automated system for identifying and dropping marginal database indexes
A method and system for managing indexes used to retrieve data from a database stored on a computer, includes determining whether any marginal indexes exist in the database, and removing one or more of the marginal indexes from the database. An index is considered marginal when: (1) it is an access...
Saved in:
| Main Author | |
|---|---|
| Format | Patent |
| Language | English |
| Published |
20.07.2010
|
| Online Access | Get full text |
Cover
Loading…
| Summary: | A method and system for managing indexes used to retrieve data from a database stored on a computer, includes determining whether any marginal indexes exist in the database, and removing one or more of the marginal indexes from the database. An index is considered marginal when: (1) it is an access path for one or more queries but the index's performance is matched or nearly matched by one or more other alternative access paths, and/or (2) its overhead has become a concern. These steps or functions are invoked by one or more statements or instructions that cause the computer to perform a Drop Index Analysis on a specified list of tables in the database for a specified workload. The Drop Index Analysis allows users to specify: (1) how many indexes to consider dropping, and/or (2) how much disk space that dropped indexes should free up. |
|---|