Re: Wrong cardinality estimates if group by (and possibly a hash join)
From: Greg Rahn <greg_at_structureddata.org>
Date: Thu, 26 Apr 2012 07:42:27 -0700
Message-ID: <CAGXkmiv2+O8Z20f9Hu=FLM35Hw122iFKseKi8qO6xtyfxy0NUw_at_mail.gmail.com>
This really isn't an issue with the group by cardinality as the output of the GBY isn't input into anything other than the final result set. The difference in the plans has to do with the forced materialization and the change of the join type.
Date: Thu, 26 Apr 2012 07:42:27 -0700
Message-ID: <CAGXkmiv2+O8Z20f9Hu=FLM35Hw122iFKseKi8qO6xtyfxy0NUw_at_mail.gmail.com>
This really isn't an issue with the group by cardinality as the output of the GBY isn't input into anything other than the final result set. The difference in the plans has to do with the forced materialization and the change of the join type.
On Thu, Apr 26, 2012 at 5:30 AM, <Laimutis.Nedzinskas_at_seb.lt> wrote:
>
> In Oracle 11.2.0.3 and 10.2.0.3 group by produces wrong cardinalities. See
> the inner hash join cardinality bellow.
>
-- Regards, Greg Rahn | blog <http://bit.ly/u9N0i8> | twitter <http://bit.ly/v733dJ> | linkedin <http://linkd.in/gregrahn> -- http://www.freelists.org/webpage/oracle-lReceived on Thu Apr 26 2012 - 09:42:27 CDT