Received: from malur.postgresql.org ([217.196.149.56])
 by arkaria.postgresql.org with esmtp (Exim 4.80) (envelope-from
 <pgsql-performance-owner+M51515=pgsql-performance-ar@postgresql.org>)
 id 1WZsdF-0003JT-LE
 for pgsql-performance@arkaria.postgresql.org; Tue, 15 Apr 2014 01:57:53 +0000
Received: from localhost ([127.0.0.1] helo=postgresql.org)
 by malur.postgresql.org with smtp (Exim 4.80) (envelope-from
 <pgsql-performance-owner+M51515=pgsql-performance-ar@postgresql.org>)
 id 1WZsdF-0000gl-1C
 for pgsql-performance@arkaria.postgresql.org; Tue, 15 Apr 2014 01:57:53 +0000
Received: from magus.postgresql.org ([2a02:c0:301:0:ffff::29])
 by malur.postgresql.org with esmtp (Exim 4.80)
 (envelope-from <ryan.johnson@cs.utoronto.ca>) id 1WZsdD-0000gK-Vp
 for pgsql-performance@postgresql.org; Tue, 15 Apr 2014 01:57:52 +0000
Received: from bureau93.ns.utoronto.ca ([128.100.132.251])
 by magus.postgresql.org with esmtp (Exim 4.80)
 (envelope-from <ryan.johnson@cs.utoronto.ca>) id 1WZsdB-0006bI-0c
 for pgsql-performance@postgresql.org; Tue, 15 Apr 2014 01:57:51 +0000
Received: from [192.168.1.114] (192-171-47-85.cpe.pppoe.ca [192.171.47.85]
 (may be forged)) (authenticated bits=0)
 by bureau93.ns.utoronto.ca (8.13.8/8.13.8) with ESMTP id s3F1vjUm032702
 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO)
 for <pgsql-performance@postgresql.org>; Mon, 14 Apr 2014 21:57:46 -0400
Message-ID: <534C9217.4010308@cs.utoronto.ca>
Date: Mon, 14 Apr 2014 21:57:43 -0400
From: Ryan Johnson <ryan.johnson@cs.utoronto.ca>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64;
 rv:17.0) Gecko/20130801 Thunderbird/17.0.8
MIME-Version: 1.0
To: pgsql-performance@postgresql.org
Subject: Re: SSI slows down over time
References: <1397484864.73109.YahooMailBasic@web122302.mail.ne1.yahoo.com>
 <534C4580.40700@cs.utoronto.ca>
In-Reply-To: <534C4580.40700@cs.utoronto.ca>
Content-Type: multipart/alternative;
 boundary="------------060800010702060406090407"
X-Pg-Spam-Score: -1.0 (-)
List-Archive: <http://archives.postgresql.org/pgsql-performance>
List-Help: <mailto:majordomo@postgresql.org?body=help>
List-ID: <pgsql-performance.postgresql.org>
List-Owner: <mailto:pgsql-performance-owner@postgresql.org>
List-Post: <mailto:pgsql-performance@postgresql.org>
List-Subscribe: <mailto:majordomo@postgresql.org?body=sub%20pgsql-performance>
List-Unsubscribe: 
 <mailto:majordomo@postgresql.org?body=unsub%20pgsql-performance>
X-Mailing-List: pgsql-performance
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

This is a multi-part message in MIME format.
--------------060800010702060406090407
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

On 14/04/2014 4:30 PM, Ryan Johnson wrote:
> FYI, here's a plot of performance over time. Each point in the graph 
> is throughput (in tps) over a 10-second measurement (~20 minutes 
> total), against a 12 WH TPC-C dataset with 24 clients banging on it. I 
> issued a pg_ctl stop/start pair between each run:


Updated result: SI definitely does not suffer any performance loss over 
time, but it's not clear what is wrong with SSI: deleting pg_serial/* 
had exactly zero impact on performance.

The two near-zero results for SSI are both cases where pg_serial/ 
ballooned to 8GB during the run (I saw two others tonight during 
testing). The one big dip for SI was a hang (the second such hang tonight).

>
> Thoughts?
> Ryan
>


--------------060800010702060406090407
Content-Type: multipart/related;
 boundary="------------010805060500050401050005"


--------------010805060500050401050005
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    <div class="moz-cite-prefix">On 14/04/2014 4:30 PM, Ryan Johnson
      wrote:<br>
    </div>
    <blockquote cite="mid:534C4580.40700@cs.utoronto.ca" type="cite">
      <meta content="text/html; charset=ISO-8859-1"
        http-equiv="Content-Type">
      FYI, here's a plot of performance over time. Each point in the
      graph is throughput (in tps) over a 10-second measurement (~20
      minutes total), against a 12 WH TPC-C dataset with 24 clients
      banging on it. I issued a pg_ctl stop/start pair between each run:<br>
    </blockquote>
    <br>
    <img src="cid:part1.02080609.08080508@cs.utoronto.ca" alt=""><br>
    Updated result: SI definitely does not suffer any performance loss
    over time, but it's not clear what is wrong with SSI: deleting
    pg_serial/* had exactly zero impact on performance. <br>
    <br>
    The two near-zero results for SSI are both cases where pg_serial/
    ballooned to 8GB during the run (I saw two others tonight during
    testing). The one big dip for SI was a hang (the second such hang
    tonight).<br>
    <br>
    <blockquote cite="mid:534C4580.40700@cs.utoronto.ca" type="cite"> <br>
      Thoughts?<br>
      Ryan<br>
      <br>
    </blockquote>
    <br>
  </body>
</html>

--------------010805060500050401050005
Content-Type: image/png;
 name="aibadhjj.png"
Content-Transfer-Encoding: base64
Content-ID: <part1.02080609.08080508@cs.utoronto.ca>
Content-Disposition: inline;
 filename="aibadhjj.png"
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--------------010805060500050401050005--

--------------060800010702060406090407--