Re: Rule-based vs Cost-based Optimizer?

In article <acc-1903951037230001_at_pfox.magna.com.au>, acc_at_pfox.dialix.oz.au (Andrew Calvin) says:
>
>In article <3k73qj$34b_at_potogold.rmii.com>, bromberg_at_hobbes (Rich Bromberg)
>wrote:
>> the cost-based optimizer in a 7.0.16 and 7.1.3 ... have observed a
>number of extremely poor 'decisions' by the optimizer.
>
>I have experienced the same problem, so we have returned to rule-based
>optimisation at all times.
>
>-- Andrew Calvin -- acc_at_pfox.dialix.oz.au --

I get mixed results on 7.1.4.
A query which I had to hack and append a NULL to force no index becuase it was really slow under 7.0.16 runs ok when COST-ALL_ROWS is chosen under 7.1.4, but really slow when COST=FIRST_ROWS (1 minute VS 9 minutes, and that for the 1st row to come back). I'll post a description including query and row counts etc. when I have time to package it up. Received on Tue Mar 21 1995 - 16:12:08 CET