pg.ddx.io  pgsql-performance@postgresql.org mailing list archive  
help / color / mirror / Atom feed
From: Andy Colson <andy@squeakycode.net>
To: Tomas Vondra <tv@fuzzy.cz>
Cc: Gerhard Wohlgenannt <wohlg@ai.wu.ac.at>
Cc: pgsql-performance@postgresql.org, Heinz-Peter Lang <heinz@langatium.net>
Cc: Gerhard Wohlgenannt <wohlg@ai.wu-wien.ac.at>
Cc: Weichselbraun, Albert <albert.weichselbraun@wu.ac.at>
Subject: Re: Sudden drop in DBb performance
Date: Mon, 05 Sep 2011 08:51:15 -0500
Message-ID: <4E64D3D3.8000205@squeakycode.net> (raw)
In-Reply-To: <33fadeb6ac1bade5ef01bd08d1fd7807.squirrel@sq.gransy.com>
References: <4E61D6B4.9090804@ai.wu.ac.at>
	<33fadeb6ac1bade5ef01bd08d1fd7807.squirrel@sq.gransy.com>

On 09/05/2011 02:48 AM, Tomas Vondra wrote:
> On 3 Září 2011, 9:26, Gerhard Wohlgenannt wrote:
>> Dear list,
>>
>> we are encountering serious performance problems with our database.
>> Queries which took around 100ms or less last week now take several
>> seconds.

>> Results of Bonnie++
>>
>> Version  1.96       ------Sequential Output------ --Sequential Input-
>> --Random-
>> Concurrency   1     -Per Chr- --Block-- -Rewrite- -Per Chr- --Block--
>> --Seeks--
>> Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP
>> /sec %CP
>> voyager         95G  1400  93 27804   3 16324   2  2925  96 41636   3
>> 374.9   4
>> Latency              7576us      233s      164s   15647us   13120ms
>> 3302ms
>> Version  1.96       ------Sequential Create------ --------Random
>> Create--------
>> voyager             -Create-- --Read--- -Delete-- -Create-- --Read---
>> -Delete--
>>                 files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
>> /sec %CP
>>                    16   141   0 +++++ +++   146   0   157   0 +++++ +++
>> 172   0
>> Latency              1020ms     128us    9148ms     598ms      37us
>> 485ms
>>
>
> That seems a bit slow ... 27MB/s for writes and 41MB/s forreads is ait
> slow with 8 drives.
>
> Tomas
>
>

Agreed, that's really slow.  A single SATA drive will get 60 MB/s.  Did you run Bonnie while the VM's were up and running?

   root@host:~# vmstat
   procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
    r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa
    0  0      0 308024 884812 40512932    0    0   464   168  353   92  4  2 84  9


Only one line?  That does not help much.  Can you run it as 'vmstat 2' and let it run while a few slow queries are performed?  Then paste all the lines?


-Andy



view thread (20+ messages)  latest in thread

Message-ID: <4E64D3D3.8000205@squeakycode.net>
Permalink:  ../4E64D3D3.8000205@squeakycode.net/
Also on:    postgresql.org/message-id/4E64D3D3.8000205@squeakycode.net

 · 

reply

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients using the --to and --cc options:
  reply via email

  To: pgsql-performance@postgresql.org
  Cc: andy@squeakycode.net, tv@fuzzy.cz, wohlg@ai.wu.ac.at, heinz@langatium.net, wohlg@ai.wu-wien.ac.at, albert.weichselbraun@wu.ac.at
  Subject: Re: Sudden drop in DBb performance
  In-Reply-To: <4E64D3D3.8000205@squeakycode.net>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox