lkml.org 
[lkml]   [1998]   [Dec]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: PATCH: Raw device IO for 2.1.131
On Thu, Dec 17, 1998 at 11:15:09AM -0500, Stefan Monnier wrote:
> >>>>> "Stephen" == Stephen Williams <steve@icarus.com> writes:
> > And what about the case the Oracle guy pointed out where they share a SCSI
> > disk between two hosts? Can that be made to work under Linux?
>
> It was an interesting case indeed, but I'm not sure about some of the details.
> It seems he presented several different cases all merged into one. Mostly
> two situations:
> - take-over in case of failure (only one machine ever accesses the drive at
> the same time). The only requirement here seems to be that hosts need
> some way to make sure that the disk data is consistent, so it's pretty much
> the same requirement as the single-host case.
> - parallel access from two hosts. I'm not quite sure why you'd want to do
> that. It seems to me that the only reason would be performance, but it
> seems that the disk and/or the SCSI bus would be the limiting factor rather
> than the host, so why have two hosts ?
>
> What am I missing ?

On the hardware side: Multi-channel RAID-Arrays, Fibrechannel, you name it.

On the economical side: It's way to expensive to have one of to database
servers sit idle most of the time. You want them both to be productive.

As always: just because you don't see it does not mean it does not exist...

Dominik Kubla
--
The text above represents my personal opinion and does not represent the
official position of my employer on the issue(s) discussed. Any official
statement by me on behalf of my employer will be marked as such.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:46    [from the cache]
©2003-2014 Jasper Spaans. hosted at Digital Ocean