Re: Slow query on table after adding 6 million rows....

In article <Bl9F5.34641$XV.1873175_at_nntp3.onemain.com>,   "Barbara Kennedy" <barbken_at_teleport.com> wrote:
> What is the explain plan? Is the index marked unusable?
> Jim
> <sbrendan_at_my-deja.com> wrote in message news:8s2b01
 $m4i$1_at_nnrp1.deja.com...
> > We have a datawarehouse database based on a 'star' schema.
> >
> > Each month we 'roll' forward the prior months data into a new months
> > partition.
> >
> > When we rolled 'Septembers' data into 'Octobers' partition query
 time on
> > just the 'October' partition slowed down and is now four times
 slower
> > than the 'September' partition for the same amount of data.
> >
> > There are five keys on the table, are are number(38).
> >
> > We have been working with Oracle support for five days and do not
 have
> > any relief yet.
> >
> > We have moved the data accross table spaces, changed index scheme,
> > changed degree of parallel, unpartition the table, all produced the
 same
> > slow results.
> >
> > Could there be a problem when copying data from one partition to
> > another?
> >
> > Looking for any ideas on what might be the problem.
> >
> > Thanks.
> > - Brendan
> >

I think Jim asked a couple of very good questions; now I will ask a coupld of dumb ones?

Did you update the statistics after the partition split?

Do you have local indexes on each partion? If so did you add a new local index for the new partion?

--
Mark D. Powell  -- The only advice that counts is the advice that
 you follow so follow your own advice --


Sent via Deja.com http://www.deja.com/
Before you buy.