From mboxrd@z Thu Jan 1 00:00:00 1970 From: Piergiorgio Sartor Subject: Re: Is this enough for us to have triple-parity RAID? Date: Fri, 20 Apr 2012 23:04:43 +0200 Message-ID: <20120420210443.GB2432@lazy.lzy> References: <4F8D228D.8060005@westcontrol.com> <20120417171609.GA2859@lazy.lzy> <4F8DD02F.1060504@westcontrol.com> <4F905690.3060301@zytor.com> <067e21e2-6f21-48a7-93a8-bb2249534155@email.android.com> <4F91B1C4.5080205@hesbynett.no> <4F91BB65.8040304@zytor.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <4F91BB65.8040304@zytor.com> Sender: linux-raid-owner@vger.kernel.org To: "H. Peter Anvin" Cc: David Brown , Alex , linux-raid@vger.kernel.org List-Id: linux-raid.ids Hi Peter, > > Yes, being a generator for GF(2^8) is a requirement for a parity > > generator (sorry for the confusing terminology here - if anyone has a > > better suggestion, please say) to be part of a 255 data disk system. > > However, being a GF generator is necessary but not sufficient - using > > parity generators (1, 2, 4, 16) will /not/ give quad parity for 255 data > > disks, even though individually each of 1, 2, 4 and 16 are generators > > for GF. [...] > It is also worth noting that there is nothing magical about GF(2^8). It > is just a reasonable tradeoff when tables are needed. I, then, ask you too. What is this story that being a generator is not enough? Is there any reference, documentation, link which can be studied in order to understand this limitation? In all RS papers I found, the only constrain put was that the Vandermonde must be constructed with generators. Not all RAID examples used them, but no paper, at least for what I understood, was limiting the generators to be also "independent". Any undestandable explanation? Thanks, bye, -- piergiorgio