Re: PK Index V Additional Column INdex

Connor McDonald <connor_mcdonald_at_yahoo.com> wrote in message news:<429323A6.365D_at_yahoo.com>...
> Sybrand Bakker wrote:
> >
> > On 23 May 2005 07:06:26 -0700, "Thiko!" <biwombi_at_hotmail.com> wrote:
> >
> > >Hi
> > >
> > >Having put some SQL through an EXPLAIN PLAN I can see that it is doing
> > >an INDEX RANGE SCAN using it's Primary Key index. This if fine.
> > >
> > >However, I tested creating an index on the column in the WHERE clause
> > >and re running the query through the EXPLAIN PLAN. It showed the
> > >explain plan to be the same but just using the new index.
> > >
> > >Would there be an advantage to having the new index created or is using
> > >the PK index in the exexution plan fine?
> > >
> > >This table has around 34000 rows in it.
> > >
> > >Many thanks for any advice!
> > >
> > >Thiko!
> >
> > Remark one: No version
> > Remark two: you did gather statistics for this index?
> > Whether the index will be used is determined by the cardinality of the
> > query expression.
> > You could set event 10051 to verify the optimizers decisions.
> >
> > --
> > Sybrand Bakker, Senior Oracle DBA
>
> I think posting about 10053 might be overkill for someone who is asking
> about whether a new index is good or bad.
>
> hth
> connor
> --
> Connor McDonald
> Co-author: "Mastering Oracle PL/SQL - Practical Solutions"
> Co-author: "Oracle Insight - Tales of the OakTable"
>
> web: http://www.oracledba.co.uk
> web: http://www.oaktable.net
> email: connor_mcdonald_at_yahoo.com
>
>
> "Semper in excremento, sole profundum qui variat."
>
> ------------------------------------------------------------

Hi Connor,

I think its more of habit, I always turn on 10053 and look through the trace
I do not want to experiment too much. The same accounts for performance problems, I always end up with a 10046 trace, I want to make sure no hidden
"newly changed behaviour" is causing my headache. Received on Tue May 24 2005 - 21:02:39 CDT