Hi,
In Oracle, there are 2 ways to do the equivalent of vacuum analyze :
* analyze table xx compute statitics
* analyze table xx estimate statistics
In the second form, you can tell on what percentage of the file you
will do your stats. This is useful to make a quick analyze on huge tables
that have a homogenous dispersion.
Is there a way to "estimate" the statistics that vacuum analyze will
use instead of "computing" them ?
--
Jean-Christophe Boggio
cat@thefreecat.org
Independant Consultant and Developer
Delphi, Linux, Perl, PostgreSQL
