Re: Query Performance issue

From: Jonathan Lewis <jlewisoracle_at_gmail.com>
Date: Tue, 29 Dec 2020 13:00:14 +0000
Message-ID: <CAGtsp8mosNyEbKFTV1+GzE2A6hw14T8kb3ZPFbn=zSw-qQwp3A_at_mail.gmail.com>



Lothar,

Almost certainly the case, but probably more complicated than a sufficiently efficient, but simple(r), rewrite.

Regards
Jonathan Lewis

On Tue, 29 Dec 2020 at 12:21, Lothar Flatz <l.flatz_at_bluewin.ch> wrote:

> Hi Mark,
>
> I wonder if Björn and mine "backbone join" technique can be used.
> Stefan describes it in short here:
>
> https://community.oracle.com/tech/apps-infra/discussion/4386851/implement-index-backbone-join-transformation-rewrite-in-cbo-code
>
> The bottom line is to rewrite the query such that the join is made only
> with the columns essentially needed.
> In my example you need to put essential columns into indexes. On an
> Exadata that might not even be necessary.
> The rest of the columns is picked up later using the rowids of the rows
> that make it into the final result.
>
> Regards
>
> Lothar
>
>

--
http://www.freelists.org/webpage/oracle-l
Received on Tue Dec 29 2020 - 14:00:14 CET

Original text of this message