Date: Wed, 3 Jun 2020 10:09:59 -0500
From: Justin Pryzby <pryzby@telsasoft.com>
To: Oleksandr Shulgin <oleksandr.shulgin@zalando.de>
Cc: pgsql-performance@lists.postgresql.org
Subject: Re: When to use PARTITION BY HASH?
Message-ID: <20200603150959.GP30144@telsasoft.com>
References: 
 <CACACo5StuUuFox+65KQHB8u+51kMUcEeQuV=XmsUv7d8VnaO7A@mail.gmail.com>
 <20200602173353.GB30144@telsasoft.com>
 <CACACo5QUUG0KoTq8uFjtW9Bpqr7d8s=dGDdjxq0R+-YfNs6KPQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: 
 <CACACo5QUUG0KoTq8uFjtW9Bpqr7d8s=dGDdjxq0R+-YfNs6KPQ@mail.gmail.com>
User-Agent: Mutt/1.9.4 (2018-02-28)
Precedence: bulk

On Wed, Jun 03, 2020 at 09:45:48AM +0200, Oleksandr Shulgin wrote:
> I see.  So it works with low cardinality in the partitioned column.  With
> high cardinality an index scan on an unpartitioned table would be
> preferable I guess.
> 
> The documentation page I've linked only contains examples around
> partitioning BY RANGE.  I believe it'd be helpful to extend it with some
> meaningful examples for LIST and HASH partitioning.

I agree.  I think it would also be useful to mention the "benefits" which
aren't likely to apply to hash partitioning.

Would you want to propose an example to include ?
Eventually it needs to be submitted as a patch to -hackers.

-- 
Justin