Skip site navigation (1) Skip section navigation (2)

Queries related to checkpoints

From: Rohan Malhotra <yourbuddyrohan(at)gmail(dot)com>
To: pgsql-general(at)postgresql(dot)org, pgsql-performance(at)postgresql(dot)org
Subject: Queries related to checkpoints
Date: 2011-07-29 13:37:15
Message-ID: CAEoM48XcWVaOU11L9CaDyMXo2kGUhA6ryPCx6kkaU1HX1zLm1Q@mail.gmail.com (view raw or flat)
Thread:
Lists: pgsql-generalpgsql-performance
Hi,

I am a Noob with db tuning and trying to analyze pg_stats_brwriter data

checkpoints_timed | checkpoints_req | buffers_checkpoint | buffers_clean |
maxwritten_clean | buffers_backend | buffers_alloc


-------------------+-----------------+--------------------+---------------+------------------+-----------------+---------------+------
            35241 |              58 |             699136 |        581839 |
          1597 |         1663650 |    2205969940


Almost all checkpoints (99.8%) that happened are because of
checkpoint_timeout passing. Is this good or should I increaase my
checkpoint_segments?
During checkpoints, 699136 8K buffers were written out which is pretty low
(less than 1MB).

buffers allocated (2205969940 8K), 1663650 times a database backend
(probably the client itself) had to write a page in order to make space for
the new allocation. Buffer allocated seems to be too high than backend
buffers.

How to read more into the data?

Regards
Rohan

Responses

pgsql-performance by date

Next:From: Kevin GrittnerDate: 2011-07-29 15:25:38
Subject: Re: [PERFORM] Queries related to checkpoints
Previous:From: Robert HaasDate: 2011-07-29 00:25:20
Subject: Re: very large record sizes and ressource usage

pgsql-general by date

Next:From: Adrian KlaverDate: 2011-07-29 13:40:31
Subject: Re: repmgr and archive_mode
Previous:From: Willy-Bas LoosDate: 2011-07-29 13:26:49
Subject: how much work_mem is my server using

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group