From: Avi Kivity <avi@redhat.com>
To: jim owens <jowens@hp.com>
Cc: Stephan von Krawczynski <skraw@ithnet.com>, linux-btrfs@vger.kernel.org
Subject: Re: Some very basic questions
Date: Wed, 22 Oct 2008 09:15:05 +0200 [thread overview]
Message-ID: <48FED2F9.1060103@redhat.com> (raw)
In-Reply-To: <48FE36BD.6020701@hp.com>
jim owens wrote:
>
> Remember that the device bandwidth is the limiter so even
> when each host has a dedicated path to the device (as in
> dual port SAS or FC), that 2nd host cuts the throughput by
> more than 1/2 with uncoordinated seeks and transfers.
That's only a problem if there is a single shared device. Since btrfs
supports multiple devices, each host could own a device set and access
from other hosts would be through the owner. You would need RDMA to get
reasonable performance and some kind of dual-porting to get high
availability. Each host could control the allocation tree for its devices.
Of course, this doesn't solve the other problems with parallel mounts.
--
I have a truly marvellous patch that fixes the bug which this
signature is too narrow to contain.
next prev parent reply other threads:[~2008-10-22 7:15 UTC|newest]
Thread overview: 79+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-10-21 17:37 Some very basic questions calin
2008-10-21 20:08 ` jim owens
2008-10-22 7:15 ` Avi Kivity [this message]
2008-10-22 14:13 ` jim owens
2008-10-22 14:25 ` Avi Kivity
-- strict thread matches above, loose matches on Subject: below --
2008-10-22 14:35 dbz
2008-10-27 15:43 ` Stephan von Krawczynski
2008-10-21 11:23 Stephan von Krawczynski
2008-10-21 12:13 ` Andi Kleen
2008-10-21 14:22 ` Stephan von Krawczynski
2008-10-21 15:34 ` jim owens
2008-10-22 11:36 ` Stephan von Krawczynski
2008-10-22 12:15 ` Avi Kivity
2008-10-22 13:03 ` Ric Wheeler
2008-10-22 13:13 ` Chris Mason
2008-10-22 13:16 ` Avi Kivity
2008-10-21 13:20 ` jim owens
2008-10-21 17:01 ` Stephan von Krawczynski
2008-10-21 17:15 ` Christoph Hellwig
2008-10-21 17:31 ` Ric Wheeler
2008-10-22 12:27 ` Stephan von Krawczynski
2008-10-22 13:15 ` Chris Mason
2008-10-22 13:27 ` Ric Wheeler
2008-10-22 14:32 ` Avi Kivity
2008-10-22 14:36 ` Chris Mason
2008-10-22 14:40 ` Avi Kivity
2008-10-22 14:46 ` Ric Wheeler
2008-10-22 14:54 ` Avi Kivity
2008-10-22 15:02 ` Ric Wheeler
2008-10-22 15:13 ` Avi Kivity
2008-10-22 15:25 ` Ric Wheeler
2008-10-22 15:33 ` Chris Mason
2008-10-22 15:43 ` Avi Kivity
2008-10-22 15:54 ` Ric Wheeler
2008-10-22 18:28 ` Avi Kivity
2008-10-22 15:39 ` Avi Kivity
2008-10-22 13:52 ` Stephan von Krawczynski
2008-10-22 15:56 ` Michel Salim
2008-10-22 16:56 ` jim owens
2008-10-23 9:47 ` Stephan von Krawczynski
2008-10-22 11:40 ` Stephan von Krawczynski
2008-10-21 13:59 ` Chris Mason
2008-10-21 16:09 ` Andi Kleen
2008-10-22 11:43 ` Stephan von Krawczynski
2008-10-21 16:27 ` Stephan von Krawczynski
2008-10-21 16:59 ` Andi Kleen
2008-10-22 11:46 ` Stephan von Krawczynski
2008-10-21 17:49 ` Chris Mason
2008-10-22 12:19 ` Stephan von Krawczynski
2008-10-22 12:48 ` Jeff Schroeder
2008-10-22 14:02 ` Stephan von Krawczynski
2008-10-22 13:50 ` Chris Mason
2008-10-22 14:04 ` Matthias Wächter
2008-10-22 14:32 ` Ric Wheeler
2008-10-22 14:44 ` jim owens
2008-10-24 8:42 ` Chris Samuel
2008-10-24 8:39 ` Chris Samuel
2008-10-21 20:54 ` Eric Anopolsky
2008-10-21 22:18 ` Ric Wheeler
2008-10-22 2:29 ` Eric Anopolsky
2008-10-22 10:42 ` Ric Wheeler
2008-10-22 10:53 ` Tejun Heo
2008-10-22 12:57 ` Ric Wheeler
2008-10-22 12:57 ` Ric Wheeler
2008-10-22 13:15 ` Tejun Heo
2008-10-22 13:19 ` Chris Mason
2008-10-22 13:38 ` Ric Wheeler
2008-10-22 13:59 ` Chris Mason
2008-10-22 14:23 ` Ric Wheeler
2008-10-22 13:23 ` Ric Wheeler
2008-10-22 16:14 ` Tejun Heo
2008-10-22 16:34 ` Ric Wheeler
2008-10-23 3:59 ` Tejun Heo
2008-10-22 18:32 ` Avi Kivity
2008-10-22 19:13 ` jim owens
2008-10-22 19:22 ` Avi Kivity
2008-10-22 19:59 ` Ric Wheeler
2008-10-22 21:31 ` Eric Anopolsky
2008-10-22 21:56 ` 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=48FED2F9.1060103@redhat.com \
--to=avi@redhat.com \
--cc=jowens@hp.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=skraw@ithnet.com \
/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.