Re: extended dynamic sampling documentation?
- From: Mladen Gogala <mgogala@xxxxxxxxxxx>
- To: ryan_gaffuri@xxxxxxxxxxx
- Date: Fri, 17 Nov 2006 10:20:18 -0500
On 11/17/2006 10:05:39 AM, ryan_gaffuri@xxxxxxxxxxx wrote:
> The documetns/articles on the web are pretty high level. anyone know of any
> documents that go into more detail. When I look at a tkprof output I see
> somewhat different dynamic sampling queries when the optimizer level is at 2
> (this is 10.2).
>
Ryan, tkprof is intended for analyzing the 10046 trace, aka "sql trace" and
nothing else. SQL trace, of course,
is next to useless when one is trying to analyze the optimizer behavior. What
you need is 10053 trace, level 1
(which actually produces more data then level 2) and the instructions which can
be found at Centrex Consulting
home page, in Jonathan's book or on Metastink, as a "case study". My favorite
is the one by Wolfgang, on Centrex
Consulting page, as this was the first one available and I re-read it gazillion
times by now. You can change dynamic
sampling at the session level and analyze the output with the different levels
of optimizer_dynamic_sampling.
This is the method one employs when dealing with dilemmas like yours.
--
Mladen Gogala
http://www.mladen-gogala.com
--
//www.freelists.org/webpage/oracle-l
Other related posts:
