lkml.org 
[lkml]   [1998]   [Jun]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Journaled FS
On Thu, Jun 18, 1998 at 02:29:40PM -0400, Adam D. Bradley wrote:
>
> > On Wed, 17 Jun 1998 12:01:37 -0400 (EDT)
> > Adam D. Bradley <artdodge@cs.bu.edu> wrote:
> >
> > > I'm working on a design for a log-structured FS with things like high
> > > availability and dynamic volumes (i.e., add/remove disks on the fly)
> > > in mind. Current working-name is "XLFS", likely to change regularly
> > > until I turn an alpha version loose on the list ;-)
> >
> > Please think about NOT adding/removing disks but instead using
> > resized volumes.
> >
> > Have a look at my Logical Volume Manager (LVM) at
> > <http://www.linux-kongress.de/nachlese98.html>
> > and
> > <ftp://ftp.msede.com/pub/linux/lvm>.
> >
> > The LVM does the job of resizing volumes at run time without
> > beeing stuck to partition limitations.
>
> The LVM is certainly a powerful tool, and a standard LFS could very
> easily take advantage of it; however, there are some kinds of
> functionality I want to play with that require explicit awareness of
> distinct physical volumes. (Consider software-variable striping and
> redundancy, for example, or ways of implementing "guaranteed disk
> bandwidth" in the style of SGI's XFS.)

The LVM is able to do provider disk with a specific allocation
sheme e.g. Raid 0/1/5. To have remote copies i thought
of implementing a "Network Block Device" which simple
might export Block device via Network (UDP ?) and then you
just include your NBD to a mirror. And there is you remote
mirror. The local mirror must be set to not read (for speed reasons)
from the network block device as this might be slower so that
only write performance gets slower through the network.

Guranteed disk performance and like should be a bit hard
to implement on the LVM layer ...

Flo
--
Florian.Lohoff@mediaWays.net +49-5241-80-7085
aka flo@mini.gt.owl.de @HOME +49-5241-470566

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu

\
 
 \ /
  Last update: 2005-03-22 13:43    [W:0.057 / U:0.092 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site