Re: tuning request and join index

From: ddf <oratune_at_msn.com>
Date: Wed, 23 Sep 2009 06:36:04 -0700 (PDT)
Message-ID: <405ac72f-574a-44fe-8bd7-8cab2ff57249_at_l34g2000vba.googlegroups.com>

On Sep 23, 5:04 am, Pif <fjs..._at_laposte.net> wrote:
> Hello,
>
> I've a request that is like:
>
> select l1.bill_id
> from file f1
>    join file f2 on f1.letter_id = f2.letter_id
>    join letter l on f1.letter_id = l.letter_id
>
> so, the file table and letter table are big ( > 100000 rows) and I
> need that execution.
> Join between both file aliases is not on ids, so there is no FK but
> there is an index.
>
> So, how can I optimize the request or the schema to have fast answer ?
> What type of index should I create and how for such case ?
>
> The only solution that my chief would like to avoid the materialized
> views...
>
> Thanks a lot for your help

There isn't enough information. How is this index defined? How are these tables defined? Where is some sample data? Which release of Oracle?

Provide at least the items requested if you want anyone to be able to provide a reasonable and usable answer.

David Fitzjarrell Received on Wed Sep 23 2009 - 08:36:04 CDT