Re: Overflow Distributions in Online 7.20
Art S. Kagel wrote:
>
> Mark D. Stock wrote:
> >
> > Art S. Kagel wrote:
> > >
> > > mc wrote:
> > > >
> > > > I am having a performance problem in Online 7.20. When I
look
> at
> > > the
> > > > distributions on the table using the "dbschema -hd" command ,
I
> > > notice that
> [SNIP]
> > That is true, but overflow bins indicate highly duplicate key
> values. If
> > most of your keys are highly duplicate, then this could impact on
> > performance. A very long and narrow B-Tree structure will incur a
> lot
> > more IO. If you have highly duplicate key values, consider adding
> > additional columns to them to increase their uniqueness.
>
> Not neccessarily so, Mark. The distributions are recorded on
> individual
> columns, high duplication on a single column does not imply high
> duplication within compound indexes only within a singleton index on
> the
> column in question. You are right that it MAY indicate a high
> duplication factor and this should be looked at but it does not of
> neccessity imply this.
It ain't necessarily so? You are right, and I thought that's what I
said. But I've been taking so much stick on my writing abilities
lately that I'm starting to get a complex! :-)
Cheers,
--
Mark.
+----------------------------------------------------------+-----------+
|Mark D. Stock - Informix SA http://www.informix.com |//////// /|
|mailto:mdstock@informix.com FAQ http://www.iiug.org |///// / //|
| +-----------------------------------+//// / ///|
| Tel: +27 11 807 0313 |If it's slow, the users complain. |/// / ////|
| Fax: +27 11 807 2594 |If it's fast, the users keep quiet.|// / /////|
|Cell: +27 83 250 2325 |Therefore, "No news: travels fast"!|/ ////////|
+----------------------+-----------------------------------+-----------+