From: Oliver Xymoron <oxymoron@waste.org>
To: James Bottomley <James.Bottomley@steeleye.com>
Cc: Lars Marowsky-Bree <lmb@suse.de>,
linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org
Subject: Re: [RFC] Multi-path IO in 2.5/2.6 ?
Date: Wed, 11 Sep 2002 16:38:11 -0500 [thread overview]
Message-ID: <20020911213811.GF31597@waste.org> (raw)
In-Reply-To: <200209111937.g8BJbfQ02442@localhost.localdomain>
On Wed, Sep 11, 2002 at 02:37:40PM -0500, James Bottomley wrote:
> lmb@suse.de said:
> > and if the device mapper could have the notion of "to what topology
> > group does this device belong to", or even "distance metric (without
> > going into further detail on what this is, as long as it is consistent
> > to the physical layer) to the current CPU" (so that the shortest path
> > in NUMA could be selected), that would be kinda cool ;-) And doesn't
> > seem too intrusive.
>
> I think I see driverfs as the solution here. Topology is deduced by examining
> certain device and HBA parameters. As long as these parameters can be exposed
> as nodes in the device directory for driverfs, a user level daemon map the
> topology and connect the paths at the top. It should even be possible to
> weight the constructed multi-paths.
>
> This solution appeals because the kernel doesn't have to dictate policy, all
> it needs to be told is what information it should be exposing and lets user
> level get on with policy determination (this is a mini version of why we
> shouldn't have network routing policy deduced and implemented by the kernel).
Not coincidentally, network routing policy _is_ multipath config in
the iSCSI or nbd case.
--
"Love the dolphins," she advised him. "Write by W.A.S.T.E.."
next prev parent reply other threads:[~2002-09-11 21:38 UTC|newest]
Thread overview: 294+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-09-09 14:57 [RFC] Multi-path IO in 2.5/2.6 ? James Bottomley
2002-09-09 16:56 ` Patrick Mansfield
2002-09-09 17:34 ` James Bottomley
2002-09-09 18:40 ` Mike Anderson
2002-09-10 13:02 ` Lars Marowsky-Bree
2002-09-10 16:03 ` Patrick Mansfield
2002-09-10 16:03 ` Patrick Mansfield
2002-09-10 16:27 ` Mike Anderson
2002-09-10 0:08 ` Patrick Mansfield
2002-09-10 7:55 ` Jeremy Higdon
2002-09-10 13:04 ` Lars Marowsky-Bree
2002-09-10 16:20 ` Patrick Mansfield
2002-09-10 16:20 ` Patrick Mansfield
2002-09-10 13:16 ` Lars Marowsky-Bree
2002-09-10 19:26 ` Patrick Mansfield
2002-09-11 14:20 ` James Bottomley
2002-09-11 19:17 ` Lars Marowsky-Bree
2002-09-11 19:17 ` Lars Marowsky-Bree
2002-09-11 19:37 ` James Bottomley
2002-09-11 19:52 ` Lars Marowsky-Bree
2002-09-11 19:52 ` Lars Marowsky-Bree
2002-09-12 1:15 ` Bernd Eckenfels
2002-09-11 21:38 ` Oliver Xymoron [this message]
2002-09-11 20:30 ` Doug Ledford
2002-09-11 21:17 ` Mike Anderson
2002-09-10 17:21 ` Patrick Mochel
2002-09-10 17:21 ` Patrick Mochel
2002-09-10 18:42 ` Patrick Mansfield
2002-09-10 19:00 ` Patrick Mochel
2002-09-10 19:00 ` Patrick Mochel
2002-09-10 19:37 ` Patrick Mansfield
2002-09-11 0:21 ` Neil Brown
[not found] <dledford@redhat.com>
2002-10-02 0:28 ` PATCH: scsi device queue depth adjustability patch Doug Ledford
2002-10-02 1:16 ` Alan Cox
2002-10-02 1:41 ` Doug Ledford
2002-10-02 13:44 ` Alan Cox
2002-10-02 21:41 ` James Bottomley
2002-10-02 22:18 ` Doug Ledford
2002-10-02 23:19 ` James Bottomley
2002-10-03 12:46 ` James Bottomley
2002-10-03 16:35 ` Doug Ledford
2002-10-04 1:40 ` Jeremy Higdon
2002-10-03 14:25 ` James Bottomley
2002-10-03 16:41 ` Doug Ledford
2002-10-03 17:00 ` James Bottomley
2002-10-16 21:35 ` scsi_scan.c question Doug Ledford
2002-10-16 21:41 ` James Bottomley
2002-10-17 0:18 ` Doug Ledford
2002-10-16 21:57 ` Patrick Mansfield
2002-10-18 15:57 ` Patrick Mansfield
2002-11-18 0:27 ` aic7xxx_biosparam Doug Ledford
2002-11-18 0:36 ` aic7xxx_biosparam J.E.J. Bottomley
2002-11-18 2:46 ` aic7xxx_biosparam Doug Ledford
2002-11-18 3:20 ` aic7xxx_biosparam J.E.J. Bottomley
2002-11-18 3:26 ` aic7xxx_biosparam Doug Ledford
2002-11-18 0:43 ` aic7xxx_biosparam Andries Brouwer
2002-11-18 2:47 ` aic7xxx_biosparam Doug Ledford
2002-11-18 0:57 ` aic7xxx_biosparam Alan Cox
2002-11-18 2:34 ` aic7xxx_biosparam Doug Ledford
2002-12-21 1:22 ` scsi_scan changes Doug Ledford
2002-12-21 1:27 ` James Bottomley
2008-10-31 1:02 ` RFC - device names and mdadm with some reference to udev greg
2008-10-31 9:18 ` Neil Brown
2008-11-02 13:52 ` Luca Berra
-- strict thread matches above, loose matches on Subject: below --
2002-11-21 15:16 [PATCH] turn scsi_allocate_device into readable code Christoph Hellwig
2002-11-21 15:36 ` Doug Ledford
2002-11-21 15:39 ` J.E.J. Bottomley
2002-11-21 15:49 ` Doug Ledford
2002-11-21 16:12 ` J.E.J. Bottomley
2002-11-21 17:08 ` [PATCH] current scsi-misc-2.5 include files Patrick Mansfield
2002-11-16 19:40 [PATCH] removel useless mod use count manipulation Christoph Hellwig
2002-11-17 2:59 ` Doug Ledford
2002-11-17 17:31 ` J.E.J. Bottomley
2002-11-17 18:14 ` Doug Ledford
2002-11-17 12:40 ` Douglas Gilbert
2002-11-17 12:48 ` Christoph Hellwig
2002-11-17 13:38 ` Douglas Gilbert
2002-11-15 20:34 [RFC][PATCH] move dma_mask into struct device J.E.J. Bottomley
2002-11-16 0:19 ` Mike Anderson
2002-11-16 14:48 ` J.E.J. Bottomley
2002-11-16 20:33 ` Patrick Mansfield
2002-11-17 15:07 ` J.E.J. Bottomley
2002-11-06 22:18 [PATCH] add request prep functions to SCSI J.E.J. Bottomley
2002-11-06 23:16 ` Doug Ledford
2002-11-06 23:43 ` J.E.J. Bottomley
2002-11-07 21:45 ` Mike Anderson
2002-11-06 4:24 [PATCH] fix 2.5 scsi queue depth setting Patrick Mansfield
2002-11-06 4:35 ` Patrick Mansfield
2002-11-06 17:15 ` J.E.J. Bottomley
2002-11-06 17:47 ` J.E.J. Bottomley
2002-11-06 18:24 ` Patrick Mansfield
2002-11-06 18:32 ` J.E.J. Bottomley
2002-11-06 18:39 ` Patrick Mansfield
2002-11-06 18:50 ` J.E.J. Bottomley
2002-11-06 19:50 ` Patrick Mansfield
2002-11-06 20:45 ` Doug Ledford
2002-11-06 21:19 ` J.E.J. Bottomley
2002-11-06 20:50 ` Doug Ledford
[not found] <patmans@us.ibm.com>
2002-10-15 16:55 ` [RFC PATCH] consolidate SCSI-2 command lun setting Patrick Mansfield
2002-10-15 20:29 ` James Bottomley
2002-10-15 22:00 ` Patrick Mansfield
2002-10-30 16:58 ` [PATCH] 2.5 current bk fix setting scsi queue depths Patrick Mansfield
2002-10-30 17:17 ` James Bottomley
2002-10-30 18:05 ` Patrick Mansfield
2002-10-31 0:44 ` James Bottomley
2002-10-21 19:34 [PATCH] get rid of ->finish method for highlevel drivers Christoph Hellwig
2002-10-21 23:58 ` James Bottomley
2002-10-22 15:48 ` James Bottomley
2002-10-22 18:43 ` Patrick Mansfield
2002-10-22 23:17 ` Mike Anderson
2002-10-22 23:30 ` Doug Ledford
2002-10-23 14:16 ` James Bottomley
2002-10-23 15:13 ` Christoph Hellwig
2002-10-24 1:36 ` Patrick Mansfield
2002-10-24 23:20 ` Willem Riede
2002-10-24 23:36 ` Christoph Hellwig
2002-10-25 0:02 ` Willem Riede
2002-10-22 7:30 ` Mike Anderson
2002-10-22 11:14 ` Christoph Hellwig
2002-10-15 18:55 [patch 2.5] ips queue depths Jeffery, David
2002-10-15 19:30 ` Dave Hansen
2002-10-15 19:47 ` Doug Ledford
2002-10-15 20:04 ` Patrick Mansfield
2002-10-15 20:52 ` Doug Ledford
2002-10-15 23:30 ` Patrick Mansfield
2002-10-15 23:56 ` Luben Tuikov
2002-10-16 2:32 ` Doug Ledford
2002-10-16 19:04 ` Patrick Mansfield
2002-10-16 20:15 ` Doug Ledford
2002-10-17 0:39 ` Luben Tuikov
2002-10-17 17:01 ` Mike Anderson
2002-10-17 21:13 ` Luben Tuikov
2002-10-15 20:10 ` Mike Anderson
2002-10-15 20:24 ` Doug Ledford
2002-10-15 20:38 ` James Bottomley
2002-10-15 22:10 ` Mike Anderson
2002-10-16 1:04 ` James Bottomley
2002-10-15 20:24 ` Mike Anderson
2002-10-15 22:46 ` Doug Ledford
2002-10-15 20:26 ` Luben Tuikov
2002-10-15 21:27 ` Patrick Mansfield
2002-10-16 0:43 ` Luben Tuikov
2002-10-21 7:28 ` Mike Anderson
2002-10-21 16:16 ` Doug Ledford
2002-10-21 16:29 ` James Bottomley
2002-10-10 15:01 [PATCH] scsi host cleanup 3/3 (driver changes) Stephen Cameron
2002-10-10 16:46 ` Mike Anderson
2002-10-10 16:59 ` James Bottomley
2002-10-10 20:05 ` Mike Anderson
2002-09-30 21:06 [PATCH] first cut at fixing unable to requeue with no outstanding commands James Bottomley
2002-09-30 23:28 ` Mike Anderson
2002-10-01 0:38 ` James Bottomley
2002-10-01 15:01 ` Patrick Mansfield
2002-10-01 15:14 ` James Bottomley
2002-10-01 16:23 ` Mike Anderson
2002-10-01 16:30 ` James Bottomley
2002-10-01 20:18 ` Inhibit auto-attach of scsi disks ? Scott Merritt
2002-10-02 0:46 ` Alan Cox
2002-10-02 1:49 ` Scott Merritt
2002-10-02 1:58 ` Doug Ledford
2002-10-02 2:45 ` Scott Merritt
2002-10-02 13:40 ` Alan Cox
2002-09-24 11:35 SCSI woes (followup) Russell King
2002-09-24 13:46 ` James Bottomley
2002-09-24 13:58 ` Russell King
2002-09-24 14:29 ` James Bottomley
2002-09-24 18:16 ` Luben Tuikov
2002-09-24 18:18 ` Patrick Mansfield
2002-09-24 19:01 ` Russell King
2002-09-24 19:08 ` Mike Anderson
2002-09-24 19:21 ` Russell King
2002-09-24 19:32 ` Patrick Mansfield
2002-09-24 20:00 ` Russell King
2002-09-24 22:23 ` Patrick Mansfield
2002-09-24 23:04 ` Russell King
2002-09-24 22:39 ` Russell King
2002-09-24 23:14 ` James Bottomley
2002-09-24 23:26 ` Mike Anderson
2002-09-24 23:31 ` James Bottomley
2002-09-24 23:56 ` Mike Anderson
2002-09-24 23:33 ` Russell King
2002-09-25 0:47 ` Mike Anderson
2002-09-25 8:45 ` Russell King
2002-09-25 2:18 ` Doug Ledford
2002-09-25 14:41 ` Russell King
2002-09-24 23:33 ` Mike Anderson
2002-09-24 23:45 ` Russell King
2002-09-25 0:08 ` Patrick Mansfield
2002-09-25 8:41 ` Russell King
2002-09-25 17:22 ` Patrick Mansfield
2002-09-25 12:46 ` Russell King
2002-09-24 17:57 ` Luben Tuikov
2002-09-24 18:39 ` Mike Anderson
2002-09-24 18:49 ` Luben Tuikov
2002-09-10 16:34 [RFC] Multi-path IO in 2.5/2.6 ? Cameron, Steve
2002-09-10 18:48 ` Alan Cox
2002-09-10 14:43 Cameron, Steve
2002-09-10 15:05 ` Alan Cox
2002-09-10 14:06 Cameron, Steve
2002-09-10 14:25 ` Alan Cox
2002-09-09 17:58 Ulrich Weigand
2002-09-09 10:49 Lars Marowsky-Bree
2002-09-09 12:23 ` Alan Cox
2002-09-10 10:30 ` Heinz J . Mauelshagen
2002-09-03 14:35 aic7xxx sets CDR offline, how to reset? James Bottomley
2002-09-03 18:23 ` Doug Ledford
2002-09-03 19:09 ` James Bottomley
2002-09-03 20:59 ` Alan Cox
2002-09-03 20:59 ` Alan Cox
2002-09-03 21:32 ` James Bottomley
2002-09-03 21:54 ` Alan Cox
2002-09-03 22:50 ` Doug Ledford
2002-09-03 23:28 ` Alan Cox
2002-09-04 7:40 ` Jeremy Higdon
2002-09-04 16:24 ` James Bottomley
2002-09-04 17:13 ` Mike Anderson
2002-09-05 9:50 ` Jeremy Higdon
2002-09-04 16:13 ` James Bottomley
2002-09-04 16:50 ` Justin T. Gibbs
2002-09-05 9:39 ` Jeremy Higdon
2002-09-05 13:35 ` Justin T. Gibbs
2002-09-05 23:56 ` Jeremy Higdon
2002-09-06 0:13 ` Justin T. Gibbs
2002-09-06 0:32 ` Jeremy Higdon
2002-09-03 21:13 ` Doug Ledford
2002-09-03 21:48 ` James Bottomley
2002-09-03 22:42 ` Doug Ledford
2002-09-03 22:52 ` Doug Ledford
2002-09-03 23:29 ` Alan Cox
2002-09-04 21:16 ` Luben Tuikov
2002-09-04 10:37 ` Andries Brouwer
2002-09-04 10:48 ` Doug Ledford
2002-09-04 11:23 ` Alan Cox
2002-09-04 16:25 ` Rogier Wolff
2002-09-04 19:34 ` Thunder from the hill
2002-09-04 19:34 ` Thunder from the hill
2002-09-03 21:24 ` Patrick Mansfield
2002-09-03 22:02 ` James Bottomley
2002-09-03 23:26 ` Alan Cox
2002-08-26 16:29 [RFC]: 64 bit LUN/Tags, dummy device in host_queue, host_lock <-> LLDD reentrancy Aron Zeh
2002-08-26 16:48 ` James Bottomley
2002-08-26 17:27 ` Mike Anderson
2002-08-26 19:00 ` James Bottomley
2002-08-26 20:57 ` Mike Anderson
2002-08-26 21:10 ` James Bottomley
2002-08-26 22:38 ` Mike Anderson
2002-08-26 22:56 ` Patrick Mansfield
2002-08-26 23:10 ` Doug Ledford
2002-08-28 14:38 ` James Bottomley
2002-08-26 21:15 ` Mike Anderson
2002-08-12 23:38 [PATCH] 2.5.31 scsi_error.c cleanup Mike Anderson
2002-08-22 14:05 ` James Bottomley
2002-08-22 16:34 ` Mike Anderson
2002-08-22 17:11 ` James Bottomley
2002-08-22 20:10 ` Mike Anderson
2002-08-05 23:53 When must the io_request_lock be held? Jamie Wellnitz
2002-08-06 17:58 ` Mukul Kotwani
2002-08-07 14:48 ` Doug Ledford
2002-08-07 15:26 ` James Bottomley
2002-08-07 16:18 ` Doug Ledford
2002-08-07 16:48 ` James Bottomley
2002-08-07 18:06 ` Mike Anderson
2002-08-07 23:17 ` James Bottomley
2002-08-08 19:28 ` Luben Tuikov
2002-08-07 16:55 ` Patrick Mansfield
2002-06-11 2:46 Proposed changes to generic blk tag for use in SCSI (1/3) James Bottomley
2002-06-11 5:50 ` Jens Axboe
2002-06-11 14:29 ` James Bottomley
2002-06-11 14:45 ` Jens Axboe
2002-06-11 16:39 ` James Bottomley
2002-06-13 21:01 ` Doug Ledford
2002-06-13 21:26 ` James Bottomley
2002-06-13 21:26 ` James Bottomley
2002-06-13 21:50 ` Doug Ledford
2002-06-13 22:09 ` James Bottomley
2002-04-08 15:18 [RFC] Persistent naming of scsi devices sullivan
2002-04-08 15:04 ` Christoph Hellwig
2002-04-08 15:59 ` Matthew Jacob
2002-04-08 16:34 ` James Bottomley
2002-04-08 18:27 ` Patrick Mansfield
2002-04-08 19:17 ` James Bottomley
2002-04-09 0:22 ` Douglas Gilbert
2002-04-09 14:35 ` sullivan
2002-04-09 14:55 ` sullivan
2002-04-08 17:51 ` Oliver Neukum
2002-04-08 18:01 ` Christoph Hellwig
2002-04-08 18:18 ` Matthew Jacob
2002-04-08 18:28 ` James Bottomley
2002-04-08 18:34 ` Matthew Jacob
2002-04-08 19:07 ` James Bottomley
2002-04-08 20:41 ` Matthew Jacob
2002-04-08 18:45 ` Tigran Aivazian
2002-04-08 20:18 ` Eddie Williams
2002-04-09 0:48 ` Kurt Garloff
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=20020911213811.GF31597@waste.org \
--to=oxymoron@waste.org \
--cc=James.Bottomley@steeleye.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=lmb@suse.de \
/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.