[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [linux-lvm] Distributed LVM/filesystem/storage

2008/6/1 Jan-Benedict Glaw <jbglaw lug-owl de>:
> On Sun, 2008-06-01 23:09:01 +0200, Hannes Dorbath <light theendofthetunnel de> wrote:
>> Jan-Benedict Glaw wrote:
>> > However I still fail to find a good approach. One initial idea was to
>> > export all physical volumes (partitions or full disks) vie NBD, join
>> > them all on a single machine into one VG/LV, create a filesystem and
>> > re-export that via NFS/SMB/...  However, all data would need to go
>> > through the net twice. While speed isn't the most important thing here
>> > (but storage is!), that's way suboptimal.
>> GlusterFS with Unify translator is exactly what you are asking for.
>> http://www.gluster.org/docs/index.php/GlusterFS
> Already had a look at that, too. "Unify" seems to not directly
> implement any measure of redundancy. So if one Brick goes down,
> everything that was there is gone. Maybe its possible to play with the
> `mirror' translator as well, but on the first look, it seems to be
> non-trivial to accomodate for a steady loss and addition of bricks.
> MfG, JBG

I haven't seen Lustre mentioned in this thread.  I think there is
kernel support, etc.  May be worth checking out. (I have not used it.)

Quoting wikipedia http://en.wikipedia.org/wiki/Lustre_(file_system)
High availability

Lustre file system high availability features include a robust
failover and recovery mechanism, making server failures and reboots
transparent. Version interoperability between successive minor
versions of the Lustre software enables a server to be upgraded by
taking it offline (or failing it over to a standby server), performing
the upgrade, and restarting it, while all active jobs continue to run,
merely experiencing a delay.

Lustre MDSs are configured as an active/passive pair, while OSSs are
typically deployed in an active/active configuration that provides
redundancy without extra overhead. Often the standby MDS is the active
MDS for another Lustre file system, so no nodes are idle in the

Greg Freemyer
Litigation Triage Solutions Specialist
First 99 Days Litigation White Paper -

The Norcross Group
The Intersection of Evidence & Technology

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]