Date: Mon, 12 Oct 2015 20:28:08 +0200
From: Andres Freund <andres@anarazel.de>
To: Shaun Thomas <bonesmoses@gmail.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: Having some problems with concurrent COPY commands
Message-ID: <20151012182808.GB10323@awork2.anarazel.de>
References: 
 <CAB78C+C9uw4hS+_k2_sR1zgjb1DyL5UonqjmW+WNno3osx=_8g@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: 
 <CAB78C+C9uw4hS+_k2_sR1zgjb1DyL5UonqjmW+WNno3osx=_8g@mail.gmail.com>
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

Hi,

On 2015-10-12 13:17:53 -0500, Shaun Thomas wrote:
> It would appear I'm running into whatever issue the xloginsert_slots patch
> tried to address, but not much discussion exists afterwards.

That patch is merged, it's just that the number of slots is
hardcoded. You can recompile postgres with different values by changing
#define NUM_XLOGINSERT_LOCKS  8
in xlog.c to a different value. A restart is enough afterwards.

> Is there any other way to
> get normal parallel COPY performance, or is that just currently impossible?
> 
> I also know 9.5 underwent a lot of locking improvements, so it might
> not be relevant. I just haven't gotten a chance to repeat my tests
> with 9.5 just yet.

Hard to say anything substantive without further information. Any chance
you could provide profiles of such a run? If yes, I can help you with
instructions. I'm just to lazy to write them up if not.

Greetings,

Andres Freund


-- 
Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance