Re: Hash partitioning.

Bruce,
Many thanks. According to PostgreSQL documentation it's only range and
list partitions are supported. My question is: when I am following your
advice, is PostgreSQL will do partitioning pruning on select? My
expectation is:
I divided my table on 128 hash partitions according let's say user_id.
When I do select * from users where user_id=? , I am expecting the
engine select from some particular partition according to my function.
The issue is critical when you working with big tables, that you can't
normally partition by range/list. The feature allow parallel select from
such table: each thread might select from his own dedicated partition.
The feature also (mainly) allow to decrease index b-tree level on
partition key column by dividing index into smaller parts.

Sincerely yours,

Yuri Levinsky, DBA
Celltick Technologies Ltd., 32 Maskit St., Herzliya 46733, Israel
Mobile: +972 54 6107703, Office: +972 9 9710239; Fax: +972 9 9710222

-----Original Message-----
From: Bruce Momjian [mailto:bruce(at)momjian(dot)us]
Sent: Tuesday, June 25, 2013 4:21 PM
To: Yuri Levinsky
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: [HACKERS] Hash partitioning.

On Tue, Jun 25, 2013 at 03:48:19PM +0300, Yuri Levinsky wrote:
> Hi,
>
> Do we have any plans to implement Hash Partitioning, maybe I missing
> this feature?

You can do it by writing your own constraint and trigger functions that
control the hashing.

--
Bruce Momjian <bruce(at)momjian(dot)us> http://momjian.us
EnterpriseDB http://enterprisedb.com

+ It's impossible for everything to be true. +

This mail was received via Mail-SeCure System.