You can store the rows in a table in either ascending or descending order. To do this, you need to define a cluster key. Using a cluster key for storing the rows in a table in ascending or descending order reduces the data I/O time for the following types of operations:

• Searching, updating, or deleting rows by specifying a range
• Searching or updating rows in cluster key sequence

Applicability

You should specify a cluster key when all the following conditions are satisfied:

• Data will be accumulated in either ascending or descending order of the key values and the data will be accessed frequently in key value sequence.
• No key values are to be changed.
• The table's rows have a fixed row size.

A cluster key is defined with the CLUSTER KEY option in the CREATE TABLE. For details about the cluster key, see the HiRDB Version 8 Installation and Design Guide.