| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
Home -> Community -> Usenet -> c.d.o.server -> Cost Based Analysis
What is the difference between
analyze table XXXX ESTIMATE statistics; analyze table XXXX COMPUTE statistics;
One of my developers ask me the question; I know the usual compute looks at every row in the table vs estimate does a percent of the table. The estimate is quick vs compute may take a while for a large table.
The table is locked while the analyze is running ie can't drop it.
I usually am faced with explaining the rule vs cost to developers but this
question has
merit. Looking at the explain plan I noticed no difference on queries;
whereas without
any analyze ie rule based may use a different explain plan.
Anyone noticed anything, I tend to tell folks to use the estimate as it
runs faster; why would the
compute be better to use.
Platform Solaris 2.5.1 / Oracle 7.3.2.3 / 50,000 rows - 2mil row tables
Thanks,
kwhitake_at_moon.jic.com
Received on Wed Feb 26 1997 - 00:00:00 CST
![]() |
![]() |