Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Mailing Lists -> Oracle-L -> RE: full-scan vs index for "small" tables

RE: full-scan vs index for "small" tables

From: <oracle-l-bounce_at_freelists.org>
Date: Wed, 28 Jun 2006 16:50:35 -0000
Message-ID: <5A8896FB2AFC5445A7DCFC5903CCA6B02CA2B1@W03856.li01r1d.lais.net>


>Mark J. Bobak
>Cases like the above are where the optimizer can go wrong. It simply
doesn't know enough about the data and it's distribution to make the correct decision.

>On Behalf Of Cary Millsap
>My experience is that most problems that people think are category II
problems are really
> category I problems in disguise. The difference can be revealed by
inspection of 10053 data.

That kind of philosophy is exactly why I moved from developers to dba. Why should I be fighting Oracle for developers salary when I can do it for dba's?
More important, my chief was not happy at all to pay developers for inspecting 10053 traces.
He happened to remember those days when correct table ordering in the from-clause and a few hints in the extreme cases would have done the job.

I understand that Oracle wants to have the same thing for OLTP and DW. The engineering problem here is that cars are for driving and plains are for flying. Let's hope no other vendor will solve this problem better. Then I can save for reeducation and concentrate on solving problems(inspecting 10053 traces for example) on Oracle platform.

Fyrirvari/Disclaimer
http://www.landsbanki.is/disclaimer

--
http://www.freelists.org/webpage/oracle-l
Received on Wed Jun 28 2006 - 11:50:35 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US