Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
![]() |
![]() |
Home -> Community -> Usenet -> c.d.o.server -> Re: SQL performance question
In article <8qo711$5tv$1_at_nnrp1.deja.com>,
terry_stjean_at_my-deja.com wrote:
> I have an SQL select statement which I would like to see if I can get
> to run faster.
> I have a transaction table with 15 million records joined to 3 master
> tables. One has 1500 records, one has 11000 records and the other
about
> 15 records.
> Besides joining these tables on the where clause, I am also selecting
> certain records.
> Any suggestions about optimizing. Does the order on the from or where
> clause matter.
>
> Terry
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>
Not too much clues, and as performance tuning is such a broad subject,
I will only make some general remarks.
If the optimizer can't decide which table to consider as driving, it
will order them from left to right (cost-based optimizer) or right-to-
left (rule-based optimizer).
If you think you have the correct order and you are using the cost-
based optimizer (ie you have statistics on your table), you might
consider using the /*+ordered*/ hint. The scenario with one table of
only 15 rows, is the typical scenario for the optimizer to go astray.
If you want to have more adequate advice, you should post the sql and
the tkprof or autotrace explain plan output.
Hth,
-- Sybrand Bakker, Oracle DBA All standard disclaimers apply ------------------------------------------------------------------------ Sent via Deja.com http://www.deja.com/ Before you buy.Received on Mon Sep 25 2000 - 14:06:56 CDT
![]() |
![]() |