Karl Reitschuster researches CBO cardinality

 

This outstanding blog has some really good insights into the behavior of Oracle cost-based SQL optimizer regarding how Oracle estimates cardinality.  Cardinality estimates are used when the optimizer makes important decisions such as how to access a table, which index to use, and what order to join tables together:

 

http://orcasoracle.bloghi.com/2006/01/14/cbo-cardinality-oh-cardinaltiy.html

 

Reitschuster first notes that when dealing with flat files (SQL*Loader, Oracle import, external tables), Oracle seems to use a cardinality=1:

 

“The strange thing is the estimated Cardinality of the FULL Table Scan. The optimizer believed that only one Compare between the MainCustomer Column and the SubCustomer Column would meet the Predicate. . .

 

CBO estimates it’s costs in for a relational database (You may never forget that). It believes that all important relational rules are kept. Specially the rule of normalization. The Relationship designed in the import table is flat and not expressible via a foreign key constraint.”

 

Reitschuster also explores the impact of CBO cardinality estimation of the choices of join methods (nested loop vs. hash join) and demonstrates the use of the CARDINALITY hint to force a hash join:

 

“The way Oracle joins depends very strongly (sic) on the expected Cardinality of the joined table.

 

Low Cardinality -> Nested Loops

 

 . . .

 

High Cardinality -> HASH JOIN

 

In the following example the true expected cardinality (sic) of the operation was set via the CARDINALITY Hint. This hint makes only sense if you now that the Cardinality is often the same which is sometimes given for external application data or for static data.”

 

Karl concludes:

 

“If you look at explain try to concentrate on the cardinality. It has a very big impact on the design of the whole execution plan. Miss estimations can lead to SQL’s with dramatically increased response times.”