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

Home -> Community -> Usenet -> c.d.o.server -> Re: How to optimize query's execution?

Re: How to optimize query's execution?

From: uf <uf_at_one.com>
Date: Tue, 30 Aug 2005 16:09:55 +0200
Message-ID: <df1pbs$ih2$1@domitilla.aioe.org>

> General rule
> When you have a query plan you are uncertain of check the correctness
> of the statistics. Regenerate the statistics if necessary using a
> large sample size. Recheck the plan.
>
> Look at the join order. If it does not appear to be the best order for
> the join to follow then rearrange the tables so that the table names
> follow the FROM clause in the order you think should be followed. Add
> an ORDEDERED hint and look at the plan to see if the CBO was able to
> follow the hint. Time test.
>
> You can also try forcing a hash join instead of a nested loops join if
> one of the tables is an order of magnitude larger than the other.
>
> Depending on your release and configuration choices several init.ora
> (spfile) parameters affect the CBO.
>
> HTH -- Mark D Powell --
>

Thanks. I only have to figure out how to do that :-) Received on Tue Aug 30 2005 - 09:09:55 CDT

Original text of this message

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