Version 5.2
|
|
Version 5.2
|
A clustering index allows data records to be clustered on pages based on the sequence of a particular index, and maintains that clustering as much as possible over the course of insert and update activity. Clustering increases the efficiency of data retrieval when it involves accessing sequential value ranges for a particular index (the clustering index). Without a clustering index, data can still be arranged on pages based on a particular index (using REORG), but there is no mechanism to maintain the arrangement as data is added and removed without additional REORGs. See the Administration Guide and the SQL Reference for details.
The CREATE INDEX statement now allows users to specify additional columns to be appended to the set of index key columns. The new INCLUDE parameter identifies columns that are included in the index, but are not part of the unique index key. This means improved performance because queries can be satisfied by accessing only the index, without reading the base table. See the Administration Guide and the SQL Reference for details.