Re: how could select id=xx so slow? - Mailing list pgsql-performance

From Yan Chunlu
Subject Re: how could select id=xx so slow?
Date
Msg-id CAOA66tEcE9wRaTLAYdS+y+L_FA86=2MZF53N+hqZ7TAuGBGnQg@mail.gmail.com
Whole thread Raw
In response to Re: how could select id=xx so slow?  (Jeff Janes <jeff.janes@gmail.com>)
List pgsql-performance
got it, thanks! without your help I really have no idea what should be fast and what supposed to be slower.

I also find "select"  involves a lot of writes:

iotop shows:

 2789 be/4 postgres      0.00 B     57.34 M  0.00 %  0.00 % postgres: goov conta 192.168.1.129(27300) SELECT

I knew that select could cause writes, but not at this magnitude....




On Fri, Jul 13, 2012 at 2:53 AM, Jeff Janes <jeff.janes@gmail.com> wrote:
On Thu, Jul 12, 2012 at 9:07 AM, Ants Aasma <ants@cybertec.at> wrote:
> On Thu, Jul 12, 2012 at 3:48 PM, Yan Chunlu <springrider@gmail.com> wrote:
>> yes the system seems overloaded, I am dealing with a simple "INSERT" but not
>> sure if it is normal that it took more time than the explain estimated:
>
> The estimated cost is in arbitrary units, its purpose is to compare
> different execution plans, not estimate time taken. So it's completely
> normal that it doesn't match actual time taken.

Right.  And to make explicit what you implied, when there is only one
to do something (like insert a row, or do maintenance on an index) it
often doesn't even attempt to cost that at all as there is no choice.
So it is not just a matter of units.

Cheers,

Jeff

pgsql-performance by date:

Previous
From: Jeff Janes
Date:
Subject: Re: DELETE vs TRUNCATE explanation
Next
From: David Kerr
Date:
Subject: slow prepare, lots of semop calls.