From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | "Tomeh, Husam" <htomeh(at)firstam(dot)com> |
Cc: | "Joshua D(dot) Drake" <jd(at)commandprompt(dot)com>, pgsql-admin(at)postgresql(dot)org |
Subject: | Re: Partitioning Option? |
Date: | 2005-02-23 19:12:26 |
Message-ID: | 21428.1109185946@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
"Tomeh, Husam" <htomeh(at)firstam(dot)com> writes:
> (I was referring to object partitioning. For instance, if I have a huge
> table with US counties as my partition key, I could create partitions
> within the same table based on the partition key (a US county for
> example). When querying, the engine will access the partition instead of
> the whole table to get the result set. This is provided in Oracle DB EE.
> So, I was wondering whether I can do similar thing in PostgreSQL since
> we're exploring PostgreSQL)
You can build it out of spare parts: either a view over a UNION ALL of
component tables, or a parent table with a bunch of inheritance
children, either way with rules to redirect insertions into the
right subtable. (With the inheritance way you could instead use
a trigger for that, which'd likely be more flexible.)
It's not going to be quite as good as the Oracle facility, mainly
because the planner has no direct understanding that it's a partitioning
setup, but you can get most of the win.
See past discussions in the mailing list archives for details.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Bruno Wolff III | 2005-02-23 19:48:42 | Re: Partitioning Option? |
Previous Message | Tomeh, Husam | 2005-02-23 18:54:50 | Re: Partitioning Option? |