Re: Strange performance response for high load times

From: "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
To: "Peter Alban" <peter(dot)alban2(at)gmail(dot)com>, "Kenneth Marshall" <ktm(at)rice(dot)edu>
Cc: <pgsql-performance(at)postgresql(dot)org>
Subject: Re: Strange performance response for high load times
Date: 2009-06-18 21:01:17
Message-ID: 4A3A64CD0200002500027DB9@gw.wicourts.gov
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

Peter Alban <peter(dot)alban2(at)gmail(dot)com> wrote:

> Why is there such a big difference ?
>
> i.e. off peek times a simple select with where (on indexed column)
> and limit taks* 40 ms* during peek times it took *2 seconds* - 50
> times slower !

If your high work_mem setting you may have been causing the OS to
discard cached data, causing disk reads where you normally get cache
hits, or even triggered swapping. Either of those can easily cause a
difference of that magnitude, or more.

-Kevin

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Bryce Ewing 2009-06-18 22:11:25 Re: Index Scan taking long time
Previous Message Peter Alban 2009-06-18 20:49:55 Re: Strange performance response for high load times