[ale] ZFS on Linux

[Jim Kinney]

On Mon, Apr 1, 2013 at 12:59 PM, Derek Atkins <warlord at mit.edu> wrote:

> Ed Cashin <ecashin at noserose.net> writes:
>
> > Thanks in advance for the notes you take.  I'm interested to hear what
> you
> > find.
> >
> > By the way, work sent me to a Nexenta course in 2010 where I learned ZFS
> stuff
> > from Richard Elling:
> >
> >   http://www.richardelling.com/
> >
> > ... and one thing that kept coming up is how users usually reach for the
> > RAID-Z options when stripes of mirrors would perform better and offer
> more
> > flexibility.  I thought that was notable.
>
> I suspect people want more storage options.  A mirror+stripe solution
> implies 100% overhead, so if you have e.g. 10 1TB disks (10TB total) you
> only get 5TB of usable storage.  OTOH if you use something like RaidZ or
> RaidZ2 then you get more storage capacity out of the same 10 drives.  So
> I think that's why people use RaidZ instead of Stripe+Mirror.
>
> Me, I'm still researching to figure out what's the best option for my
> future use.  I'd like to be able to add more drives to expand the array.
> I'd like to be able to replace the drives (potentially with larger
> drives) and have the system expand the array when possible.  I'd like to
> be able to rebalance the system as I add more storage over time.  But
> I'd also like to have redundancy such that I can theoretically lose more
> than one drive and still survive (which would be a major issue if I had
> an 20 or 24-drives in use).  So I'm not sure if a Stripe+Mirror or
> "RaidZ" or potentially a set of striped RaidZs would be better for me.
>

" You can create a double-parity or triple-parity RAID-Z configuration by
using the raidz2 or
raidz3 keyword when creating the pool" - that handles the multi-drive loss
but the rest may be still in the fantasy/dream filesystem stages I think :-(

I know a combination of RAID and LVM2 will provide much filesystem
flexibility and redundancy. It looks like ZFS pools will do similar tricks.
Add new drives, raidz them and add that to a pool (maybe - still reading).

Replacing small drives with larger drives without data loss or space loss
would be fantastic. Rich now, the ONLY way I've every seen for this is to
partition the new drive and add slices, ignore the unused space until all
drive in array are upgraded then create new array with unused space. Not a
good way.

>
> I wonder how long until we see a LinNAS (ala FreeNAS but built on
> Linux)?
>
> -derek
>
> --
>        Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
>        Member, MIT Student Information Processing Board  (SIPB)
>        URL: http://web.mit.edu/warlord/    PP-ASEL-IA     N1NWH
>        warlord at MIT.EDU                        PGP key available
>
> _______________________________________________
> Ale mailing list
> Ale at ale.org
> http://mail.ale.org/mailman/listinfo/ale
> See JOBS, ANNOUNCE and SCHOOLS lists at
> http://mail.ale.org/mailman/listinfo
>



-- 
-- 
James P. Kinney III
*
*Every time you stop a school, you will have to build a jail. What you gain
at one end you lose at the other. It's like feeding a dog on his own tail.
It won't fatten the dog.
- Speech 11/23/1900 Mark Twain
*
http://electjimkinney.org
http://heretothereideas.blogspot.com/
*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ale.org/pipermail/ale/attachments/20130401/517eadb6/attachment.html>