Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
![]() |
![]() |
Home -> Community -> Mailing Lists -> Oracle-L -> how to detect skew in a column?
Greetings,
I am trying to write some perl to query the database and spit out list of candidate columns for histograms. I know that dbms_stats has the SKEWONLY method opt, but that also builds histograms automatically, I don't want to do that, just want candidates that I can look investigate. I've traced a dbms_stats session with method opt 'SKEW ONLY', so I think I see how dbms_stats detects skew, but I'm still investigating.
Any ideas how to identify columns that have skew in them? Besides the eyeball approach: select count(*),val from table group by val and examine the result set.
josh
![]() |
![]() |