Re: Index location patch for review

I am very new to this mailinglist so I apologize if I start talking early but
I've been working as a sysadmin and that kind of problems for a long while
now and my suggestion is that it is a start but I think that we should aim a
little higher than this and use something more like the Oracle approach
instead. Where they introduce an abstraction layer in the form of a
tablespace. And this tablespace is then referenced from the create table or
create index instead.
eg:
table -> tablespace -> path to physical storage
index -> tablespace -> path to physical storage

Advantages:
Changes can be done to storage whithout need to change create scripts for db,
tables and so on.
Designers can specify in which tablespace tables/indexes should reside based
on usage.
Sysadmins can work with tablespaces and change paths without changing
anything in the database/table/index definitions.

The alternative is symlinks to distribute the load and that is not a pretty
sight dba-wise.

Hope you can bare with me on this, since I think it is an very important
issue.
I'm unfortunately not a fast coder yet (but I'm getting faster :-) ). But I
could start writing a spec if someone is interrested.

Bruce Momjian wrote:

> > Hi all,
> >
> > Attached is a patch that adds support for specifying a location for
> > indexes via the "create database" command.
> >
> > I believe this patch is complete, but it is my first .
>
> This patch allows index locations to be specified as different from data
> locations. Is this a feature direction we want to go in? Comments?
>
> --
> Bruce Momjian | http://candle.pha.pa.us
> pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 853-3000
> + If your life is a hard drive, | 830 Blythe Avenue
> + Christ can be your backup. | Drexel Hill, Pennsylvania 19026
>
> ---------------------------(end of broadcast)---------------------------
> TIP 6: Have you searched our list archives?
>
> http://www.postgresql.org/search.mpl