So far only had to use RBO, now we are moving towards using CBO. However none of the documents really talk in detail about Histograms, the goods, bads where and where not to use.
I have read that HISTOGRAMS are good for columns that are non unique and that have many duplicate values. While DSS systems have more occurances of these kind of columns, OLTP databases could also have certain columns like status codes or state codes that have duplicate values.
Is it OK to have histograms generated for all columns irrespective of the type of system the database supports? and pros and cons?
Could some one point me to a good white paper or material that would discuss this?
TIA
Menon
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now
