From: Helge Hafting <helge.hafting@aitel.hist.no>
To: Jens Axboe <jens.axboe@oracle.com>
Cc: Jakob Oestergaard <jakob@unthought.net>,
Arjan van de Ven <arjan@infradead.org>,
"Phetteplace, Thad (GE Healthcare,
consultant)" <Thad.Phetteplace@ge.com>,
linux-kernel@vger.kernel.org
Subject: Re: Bandwidth Allocations under CFQ I/O Scheduler
Date: Wed, 18 Oct 2006 13:00:34 +0200 [thread overview]
Message-ID: <45360952.5020307@aitel.hist.no> (raw)
In-Reply-To: <20061018095125.GE24452@kernel.dk>
Jens Axboe wrote:
> While that may make some sense internally, the exported interface would
> never be workable like that. It needs to be simple, "give me foo kb/sec
> with max latency bar for this file", with an access pattern or assumed
> sequential io.
>
> Nobody speaks of iops/sec except some silly benchmark programs. I know
> that you are describing pseudo-iops, but it still doesn't make it more
> clear.
> Things aren't as simple
>
How about "give me 10% of total io capacity?" People understand
this, and the io scheduler can then guarantee this by ensuring
that the process gets 1 out of 10 io requests as long as it
keeps submitting enough.
The admin can then set a reasonable percentage depending on
the machine's capacity.
Helge Hafting
next prev parent reply other threads:[~2006-10-18 11:04 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-10-16 20:46 Bandwidth Allocations under CFQ I/O Scheduler Phetteplace, Thad (GE Healthcare, consultant)
2006-10-17 1:24 ` Arjan van de Ven
2006-10-17 13:23 ` Jens Axboe
2006-10-17 14:37 ` Ric Wheeler
2006-10-17 14:47 ` Jens Axboe
2006-10-17 14:46 ` Phetteplace, Thad (GE Healthcare, consultant)
2006-10-18 8:00 ` Jakob Oestergaard
2006-10-18 9:40 ` Arjan van de Ven
2006-10-18 11:30 ` Jakob Oestergaard
2006-10-18 11:49 ` Jens Axboe
2006-10-18 12:23 ` Jakob Oestergaard
2006-10-18 12:42 ` Alan Cox
2006-10-18 12:44 ` Jens Axboe
2006-10-18 12:55 ` Nick Piggin
2006-10-18 13:04 ` Jens Axboe
2006-10-18 13:39 ` Jakob Oestergaard
2006-10-18 13:51 ` Paulo Marques
2006-10-19 12:22 ` Jens Axboe
2006-10-18 13:37 ` Jakob Oestergaard
2006-10-18 12:44 ` Jakob Oestergaard
2006-10-18 12:42 ` Jens Axboe
2006-10-18 13:35 ` Jakob Oestergaard
2006-10-18 9:51 ` Jens Axboe
2006-10-18 11:00 ` Helge Hafting [this message]
2006-10-18 11:14 ` Jens Axboe
2006-10-18 11:23 ` Ric Wheeler
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=45360952.5020307@aitel.hist.no \
--to=helge.hafting@aitel.hist.no \
--cc=Thad.Phetteplace@ge.com \
--cc=arjan@infradead.org \
--cc=jakob@unthought.net \
--cc=jens.axboe@oracle.com \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.