Re: Statspack ratios help

That is possible.

At one extreme, it is possible that the figures show that 5.5 times per second, a new session logs in and executes about 200 statements (1085/5.5) that have recently been executed by another session.

You would need to check the statistic
"cursor authentications" to get an idea of how close to the truth this was. The first time one session issues a parse call that finds a valid cursor created by another session, the parse call is recorded as a 'cursor authentication' - so the 5.5 x 200 scenario would show a very high value for that statistic.

Regards

Jonathan Lewis
http://www.oracle.com/technology/community/oracle_ace/ace1.html#lewis

The Co-operative Oracle Users' FAQ
http://www.jlcomp.demon.co.uk/faq/ind_faq.html

Cost Based Oracle: Fundamentals
http://www.jlcomp.demon.co.uk/cbo_book/ind_book.html

• Original Message ----- From: "Igor Neyman" <ineyman_at_perceptron.com> To: <jonathan_at_jlcomp.demon.co.uk>; <oracle-l_at_freelists.org> Sent: Wednesday, June 07, 2006 2:08 PM Subject: RE: Statspack ratios help

Could it be, that lots of "soft parses" are due to the application that constantly connects/ disconnects (5.54 Logons per second), but still executes the same sql in different sessions?

-----Original Message-----
From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Jonathan Lewis Sent: Wednesday, June 07, 2006 3:25 AM
To: oracle-l_at_freelists.org
Subject: Re: Statspack ratios help

Execute to Parse %: 9.62

    This is defined by

        100 * (executes - parses) / executes     In your case (using the per-second figures)

        (1085.95 - 981.47) / 1085.95

I don't tend to look at the ratios as they lose scale, whereas the absolute figures give you some idea of the possible size of the problem.

--
http://www.freelists.org/webpage/oracle-l