AW: Reduce latch row cache objects with event 10089
Date: Tue, 17 Mar 2015 15:29:51 +0000
Message-ID: <09011014EB621E4CBC2536B62A1B6407C88B2DBE_at_smxc002.trivadis.com>
Hallo Jonathan,
thank you very much for your answer.We already use the previous options (all_rows,binds, cursor_sharing=force) I have also tried to set the optimizer_features for single statement to 10.2.0.4, which resulted in bad plan.
Best Regards,
Petr
Von: oracle-l-bounce_at_freelists.org [oracle-l-bounce_at_freelists.org]" im Auftrag von "Jonathan Lewis [jonathan_at_jlcomp.demon.co.uk] Gesendet: Dienstag, 17. März 2015 15:41
An: oracle-l_at_freelists.org
Betreff: RE: Reduce latch row cache objects with event 10089
Take note that it's the last option on the list - there's a good reason for that.
You may find that some current execution plans change (which may be for the worse) if you set this event, especially if you've set the optimizer_mode to a low first_rows(N). This is because the alphabetic tie-break will no longer apply for matching costs, and you may find Oracle changing its choice of index for accessing a table, and the order of accessing rows can make a big difference to the efficiency.
Regards
Jonathan Lewis
http://jonathanlewis.wordpress.com
_at_jloracle
From: oracle-l-bounce_at_freelists.org [oracle-l-bounce_at_freelists.org] on behalf of Petr Novak [Petr.Novak_at_trivadis.com] Sent: 17 March 2015 14:16
To: oracle-l_at_freelists.org
Subject: Reduce latch row cache objects with event 10089
Hallo,
according Metalink Note 1485410.1
event 10089 could be set. Does anybody has experience with it ?
Best Regards,
Petr--
http://www.freelists.org/webpage/oracle-l
-- http://www.freelists.org/webpage/oracle-l-- http://www.freelists.org/webpage/oracle-lReceived on Tue Mar 17 2015 - 16:29:51 CET