Date: Thu, 23 Oct 2014 06:24:00 -0700 (PDT)
From: pinker <pinker@onet.eu>
To: pgsql-performance@postgresql.org
Message-ID: <1414070640656-5824026.post@n5.nabble.com>
Subject: Checkpoints tuning
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

I have saved data from pg_stat_bgwriter view following Greg Smith's advice
from his book:
select now(),* from pg_stat_bgwriter; 

and then aggregated the data with query from his book as well.

checkpoint segments was first 30 and next day I have increased it to 200,
and results has changed:


<http://postgresql.1045698.n5.nabble.com/file/n5824026/Auswahl_235.png> 


now percent of checkpoints required because of number of segments is bigger
and backend writer share is also too high- I assume it's not what should
happen.
I'm not sure how to interpret correlation between allocation and written
data?
The bigger amount of data written per sec is a good sign?
 

--
View this message in context: http://postgresql.1045698.n5.nabble.com/Checkpoints-tuning-tp5824026.html
Sent from the PostgreSQL - performance mailing list archive at Nabble.com.


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