Date: Sun, 27 Jan 2019 10:04:06 -0600
From: Justin Pryzby <pryzby@telsasoft.com>
To: Saurabh Nanda <saurabhnanda@gmail.com>
Cc: Jeff Janes <jeff.janes@gmail.com>,
	pgsql-performance@lists.postgresql.org
Subject: Re: Benchmarking: How to identify bottleneck (limiting factor) and
 achieve "linear scalability"?
Message-ID: <20190127160406.GL13248@telsasoft.com>
References: 
 <CAPz=2oGdmvirLNX5kys+uiY7LKzCP4sTiXXob39qq6eDkEuk2Q@mail.gmail.com>
 <CAMkU=1yNvhDow9QvKeG5nzv2fDG9VXJ3gi8JbX2h0t8_MY1hBA@mail.gmail.com>
 <CAPz=2oF8C-JAny9eM_gNkzQDj1CDjGfoQ17rcHm5-N-hOFAhUg@mail.gmail.com>
 <CAPz=2oE7HBJo59ANsYoEwyJU3AbjXU1opu_p3MuYZbcja_LCAg@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: 
 <CAPz=2oE7HBJo59ANsYoEwyJU3AbjXU1opu_p3MuYZbcja_LCAg@mail.gmail.com>
User-Agent: Mutt/1.5.23 (2014-03-12)
Precedence: bulk

On Sun, Jan 27, 2019 at 01:09:16PM +0530, Saurabh Nanda wrote:
> It seems that PGOPTIONS="-c synchronous_commit=off" has a significant
> impact. However, I still can not understand why the TPS for the optimised
> case is LOWER than the default for higher concurrency levels!

Do you know which of the settings is causing lower TPS ?

I suggest to check shared_buffers.

If you haven't done it, disabling THP and KSM can resolve performance issues,
esp. with large RAM like shared_buffers, at least with older kernels.
https://www.postgresql.org/message-id/20170718180152.GE17566%40telsasoft.com

Justin