Re: Challenge: Partitioning is a wrong idea

xhos..._at_gmail.com wrote:
> mikharakiri_nospaum_at_yahoo.com wrote:
> > I'm returning to this topic because of rumors that "interval
> > partitiong" would be the next new shiny administrative toy.
>
> What do you mean by "interval partitioning"? I've heard that in the
> context of optimal organization for spatial data structures, but not
in
> the context of table partitioning.

Automatic range partitioning.

> > It is undeniable that partitioning concept introduces extra
> > complications. You have to be aware of many extra technicalities:
what
> > is partition prunning, what is partition wise join, etc.
>
> You only need to be aware of those things if you need to micromanage
the
> CBO.
Ideally, you never look under the optimizer hood. In practice, you do. Partitioning brings more complexity and, therefore, more bugs. As if optimization isn't quite difficult already.

> Unfortunately, reorganizing your "normal" table into whatever form is
most
> conducive to the single particular task you need to do at the moment
is not
> feasible for large tables.

Spell it out: what kind of table reorganization do you need. Dropping a partition is a straw man. Anything else? Received on Thu Apr 07 2005 - 18:00:09 CDT