Re: index organized tables use case

On 12/12/07, Enrico Sirola <enrico(dot)sirola(at)gmail(dot)com> wrote:
> Hello Isak,
>
> Isak Hansen ha scritto:
>
> > Have a look at the cluster operation;
> > <http://www.postgresql.org/docs/8.3/static/sql-cluster.html>.
> >
> > AFAIK it does lock & duplicate the whole table during reordering,
> > which may or may not be an issue for you.
>

Sorry Enrico and list, the respond-to setting on this list gets me every time..

> thanks for the reply; I was aware about this option, anyway I think
> probably it won't be practical: the table is very big and this exclusive
> lock would probably be a pain. I think probably I will start with the
> present relation and then redesign the application at a second stage

This is a long shot, but if the table is huge you could also consider
partitioning. That really depends on the contents of 'code', though..

See <http://www.postgresql.org/docs/8.3/static/ddl-partitioning.html>.

Isak

> Thanks,
> e.
>