From: Matt Clark <matt@ymogen.net>
To: John Meinel <john@johnmeinel.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Curtis Zinzilieta <curtisz@norchemlab.com>
Cc: Anjan Dave <adave@vantage.com>
Cc: Rod Taylor <pg@rbt.ca>
Cc: Postgresql Performance <pgsql-performance@postgresql.org>
Subject: Re: can't handle large number of INSERT/UPDATEs
Date: Wed, 27 Oct 2004 07:09:21 +0100
Message-ID: <417F3B91.9080002@ymogen.net> (raw)
In-Reply-To: <417F2181.4000801@johnmeinel.com>
References: <Pine.LNX.4.44.0410261557560.23730-100000@hyperion2.norchemlab.com>
<15812.1098847291@sss.pgh.pa.us>
<417F2181.4000801@johnmeinel.com>
>> and certainly anyone who's been around a computer more than a week or
>> two knows which direction "in" and "out" are customarily seen from.
>>
>> regards, tom lane
>>
>
Apparently not whoever wrote the man page that everyone copied ;-)
> Interesting. I checked this on several machines. They actually say
> different things.
>
> Redhat 9- bi: Blocks sent to a block device (blocks/s).
> Latest Cygwin- bi: Blocks sent to a block device (blocks/s).
> Redhat 7.x- bi: Blocks sent to a block device (blocks/s).
> Redhat AS3- bi: blocks sent out to a block device (in blocks/s)
>
> I would say that I probably agree, things should be relative to the
> cpu. However, it doesn't seem to be something that was universally
> agreed upon. Or maybe the man-pages were all wrong, and only got
> updated recently.
>
Looks like the man pages are wrong, for RH7.3 at least. It says bi is
'blocks written', but an actual test like 'dd if=/dev/zero of=/tmp/test
bs=1024 count=16384' on an otherwise nearly idle RH7.3 box gives:
procs memory swap io
system cpu
r b w swpd free buff cache si so bi bo in cs us
sy id
0 0 0 75936 474704 230452 953580 0 0 0 0 106 2527 0
0 99
0 0 0 75936 474704 230452 953580 0 0 0 16512 376 2572
0 2 98
0 0 0 75936 474704 230452 953580 0 0 0 0 105 2537
0 0 100
Which is in line with bo being 'blocks written'.
M
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: matt@ymogen.net, john@johnmeinel.com, tgl@sss.pgh.pa.us, curtisz@norchemlab.com, adave@vantage.com, pg@rbt.ca
Subject: Re: can't handle large number of INSERT/UPDATEs
In-Reply-To: <417F3B91.9080002@ymogen.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