From mboxrd@z Thu Jan  1 00:00:00 1970
From: Piergiorgio Sartor <piergiorgio.sartor@nexgo.de>
Subject: Re: Is this enough for us to have triple-parity RAID?
Date: Tue, 17 Apr 2012 22:54:05 +0200
Message-ID: <20120417205405.GA8311@lazy.lzy>
References: <CAMqP1GVnOErhxj8RV=KFtzHHEJqtguEd1L_vmsS=AAi_rAPoVg@mail.gmail.com>
 <4F8D228D.8060005@westcontrol.com>
 <20120417171609.GA2859@lazy.lzy>
 <4F8DD02F.1060504@westcontrol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Return-path: <linux-raid-owner@vger.kernel.org>
Content-Disposition: inline
In-Reply-To: <4F8DD02F.1060504@westcontrol.com>
Sender: linux-raid-owner@vger.kernel.org
To: David Brown <david@westcontrol.com>
Cc: Piergiorgio Sartor <piergiorgio.sartor@nexgo.de>, linux-raid@vger.kernel.org
List-Id: linux-raid.ids

Hi David,

On Tue, Apr 17, 2012 at 10:18:55PM +0200, David Brown wrote:
[some stuff removed]

I think that we must understand a bit better
systematic RS.

The ideal case would be to have a RAID system
where, while adding a disk, the user can decide
if it (the disk) would be for data extension or
for redundancy (parity) extension.

This should be always possible and, in the first
place, performance should not be the target.

As mentioned before, par2 does it without too
many issues, so I fail to see why three parities
should be the limit.

Actually, I'm trying to catch some RS expert in
order to get an explanation (Q&A) on how the
whole thing works. Unfortunately experts seem to
be quite sluggish... :-)

bye,

-- 

piergiorgio