From mboxrd@z Thu Jan  1 00:00:00 1970
From: Colonel Hell <amritanshu@gmail.com>
Subject: RAID6 Query
Date: Tue, 16 Aug 2005 22:30:06 +0800
Message-ID: <4ecc821305081607304a7c4da0@mail.gmail.com>
References: <20050717182650.24540.patches@notabene>
	 <E1Du4Uu-0006O7-84@notabene.cse.unsw.edu.au>
	 <009001c58ac8$9ab25d40$0400a8c0@LocalHost>
	 <17114.55335.687696.686786@cse.unsw.edu.au>
	 <03e501c5a120$f3a79a00$0400a8c0@LocalHost>
	 <17151.60931.26972.713074@cse.unsw.edu.au>
	 <011101c5a187$3bc8cf00$0400a8c0@LocalHost>
	 <016001c5a26a$05ad3e40$0400a8c0@LocalHost>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7BIT
Return-path: <linux-raid-owner@vger.kernel.org>
In-Reply-To: <016001c5a26a$05ad3e40$0400a8c0@LocalHost>
Content-Disposition: inline
Sender: linux-raid-owner@vger.kernel.org
To: linux-raid@vger.kernel.org
List-Id: linux-raid.ids

Hi,
I just went thru a couple of papers describing RAID6. 
I dunno how relevant this discussion grp is for the qry ...but here I go :) ...
I couldnt figure out why is P+Q configuration better over P+q' where
q' == P. What I mean is instead of calculating a new checksum (thru a
lot of GF theory etc) just store the parity block (P)again. In this
case as well we have the same amount of fault tolerance or not
:-s  ...   
Let me know,

here are the links which I went thru.

http://kernel.org/pub/linux/kernel/people/hpa/raid6.pdf

http://www.intel.com/design/storage/papers/308122.htm

Regards,
Amritanshu.