Dan

• Jul 9th, 2005

 

A cluster index is needed in support of a cluster. One is used to allow the other to exist. Much like we need an index in support of a primary key. They are not the same things. 
 
Clusters are useful in the database to store related pieces of information from more then 1 table in the same physical database block. It in effect stores data "prejoined". I can use this technique to store all of the data from the DEPT  
and EMP table for a given DEPTNO on the same block so that all employees of department 10 as well as the department 10 master record are all physically stored on the same exact block. When I go to "join" this data -- it is already  
done for me -- in a very few IOs I can get all of the data I need. Using conventional tables, this data could be scattered onto many dozens of blocks. 
 
read the oracle documentation, it gives examples of creating a cluster, the cluster key index, and the tables in the cluster. 
 
Also, with a hash cluster there is no index by definition so no syntax to inspect. We hash the cluster key to determine where the data goes -- we do not index it. the data is the index. 
 
 
Clusters are useful when you want data with the same cluster key values to be  
physically stored near eachother. 

imran44

 Profile Answers by imran44 

• Jan 9th, 2009

 

The related columns of the tables in a cluster are called the cluster key. The cluster key is indexed so that rows of the cluster can be retrieved with a minimum amount of I/O.