Message-ID: <5239D477.7040602@optionshouse.com>
Date: Wed, 18 Sep 2013 11:27:35 -0500
From: Shaun Thomas <sthomas@optionshouse.com>
Reply-To: <sthomas@optionshouse.com>
Organization: Peak6
User-Agent: Mozilla/5.0 (X11; Linux i686 on x86_64;
 rv:23.0) Gecko/20100101 Thunderbird/23.0
MIME-Version: 1.0
To: <fburgess@radiantblue.com>, <pgsql-performance@postgresql.org>
Subject: Re: autovacuum and dead tuples
References: 
 <20130918084242.5a830134ae84016b0174832fdc1a3173.28709778ba.wbe@email11.secureserver.net>
In-Reply-To: 
 <20130918084242.5a830134ae84016b0174832fdc1a3173.28709778ba.wbe@email11.secureserver.net>
Content-Type: text/plain; charset="UTF-8"; format=flowed
content-transfer-encoding: 7bit
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

On 09/18/2013 10:42 AM, fburgess@radiantblue.com wrote:

> My colleague is under the impression that dead tuples are only cleaned up via vacuum
> full only, while I suggested that the autovaccum process was cleaning up
> these dead tuples. Is this true?

You are correct. Only VACUUM FULL (or CLUSTER) physically removes dead 
tuples from the table, but a regular VACUUM enters them into the free 
space map for reuse, so they wouldn't show up in the dead_tuple_count 
column in pgstattuple. It's possible your colleague was confused by the 
physical removal versus reassignment.

Keep in mind that the dead tuples are still in the table, but reusable. 
The free_space column and free_percent is a better description of table 
bloat from data turnover cleaned up by autovacuum.

-- 
Shaun Thomas
OptionsHouse | 141 W. Jackson Blvd. | Suite 500 | Chicago IL, 60604
312-676-8870
sthomas@optionshouse.com

______________________________________________

See http://www.peak6.com/email_disclaimer/ for terms and conditions related to this email


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