[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4A43612F.2000100@anonymous.org.uk>
Date: Thu, 25 Jun 2009 12:36:15 +0100
From: John Robinson <john.robinson@...nymous.org.uk>
To: NeilBrown <neilb@...e.de>
CC: "Martin K. Petersen" <martin.petersen@...cle.com>,
Mike Snitzer <snitzer@...hat.com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Alasdair G Kergon <agk@...hat.com>, jens.axboe@...cle.com,
linux-scsi@...r.kernel.org, linux-kernel@...r.kernel.org,
linux-raid@...r.kernel.org, linux-ide@...r.kernel.org,
linux-fsdevel@...r.kernel.org,
device-mapper development <dm-devel@...hat.com>
Subject: Re: [dm-devel] REQUEST for new 'topology' metrics to be moved out
of the 'queue' sysfs directory.
On 25/06/2009 12:07, NeilBrown wrote:
[...]
> stripe-width is only really needed on raid4/5/6 as it is aimed at
> avoiding read-modify-write, so it would be the stripe size, which would
> be minimum_io_size.
[...]
> stride-size is used for raid0 or raid4, or the "Asymmetric" raid5 layouts.
> It allows ext3 to stagger which drive certain 'hot' data structures are
> on. The current metrics don't allow for that at all.
> I'm not saying they should, and I'm not sure how they could. But it
> is sad.
Even sadder, when a raid 0/4/5/6 is reshaped over more discs (and
probably other scenarios outwith md), both stripe-width and stride-size
change. Is there any prospect this new stacking could give us the
opportunity to tell our client (LVM, filesystem, whatever) about the
change, or that they'll be able to take advantage of it?
Cheers,
John.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists