Re: Fragmented vs. Contiguous vs. Single-extent segments

I don't agree with you. The Cost Based optimizer will tell you (if db_file_multiblock_read_count has been set appropiately) in much more cases a full table scan in combination with a hash join is the most efficient solution.

Regards,

Sybrand Bakker, Oracle DBA

"Nick P" <nick_pope_at_hotmail.com> wrote in message news:tdgts94f4k8ic9_at_xo.supernews.co.uk...
> Interesting, but not a 'real life' experiment. Very few systems will do
 full
> table scans. I would more interested to see how access speeds using
 indexes
> would be affected.
>
> "Nuno Souto" <nsouto_at_nsw.bigpond.net.au.nospam> wrote in message
> news:3ad83759.4426325_at_news-server...
> > On Sat, 14 Apr 2001 10:39:04 +0100, millerp_at_sql.org.uk (Paul Miller)
> > wrote:
> >
> > >
> > >Your comments are welcome.
> > >
> >
> > Doesn't surprise me at all. I've been saying that defragging a
> > database is completely useless since V7. And in V6, it wasn't really
> > necessary provided the correct init.ora parameters were set.
> >
> > But I'll let all the industrious dbas out there who defrag daily
> > religiously continue to do their thing. It looks good in the resumes.
> >
> > BTW, congratulations on your effort. Looks like you did a good job of
> > pointing out the real problem that may be caused by fragmentation.
> >
> >
> > Cheers
> > Nuno Souto
> > nsouto_at_bigpond.net.au.nospam
> > http://www.users.bigpond.net.au/the_Den/index.html
>
>
Received on Sat Apr 14 2001 - 12:26:36 CDT