Re: Has anyone experimented with OPTIMZER_INDEX_COST

From: Nuno Souto <nsouto_at_nsw.bigpond.net.au.nospam>
Date: Thu, 31 May 2001 11:08:33 GMT
Message-ID: <3b1623e3.1919555@news-server>

On Thu, 31 May 2001 00:23:48 GMT, "David Sisk" <davesisk_at_ipass.net> wrote:

>
>I tried leaving OPTIMIZER_GOAL=CHOOSE and setting OPTIMIZER_INDEX_COST_ADJ=
>some low value, and the optimizer chose the same execution plan, but it took
>several seconds to do so. I'm just wondering if anyone else has noticed
>this unusual (or it seems so to me) behavior.
>

Yes. I had to set it to FIRST_ROWS to get rid of this. AND still keep the low-value of the other parameter. But I admit my case was weird to start with. V8.0.6

One side effect of FIRST_ROWS and low OPTIMIZE_INDEX_COST_ADJ: your export dictionary views become exceedingly slow. I had to modify them to force RULE and re-compile the lot. Now works fine. No, I don't have stats on SYS tables at all!

Cheers
Nuno Souto
nsouto_at_bigpond.net.au.nospam
http://www.users.bigpond.net.au/the_Den/index.html Received on Thu May 31 2001 - 06:08:33 CDT

Re: Has anyone experimented with OPTIMZER_INDEX_COST_ADJ???