Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932498AbXHDREJ (ORCPT ); Sat, 4 Aug 2007 13:04:09 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1764167AbXHDRDx (ORCPT ); Sat, 4 Aug 2007 13:03:53 -0400 Received: from relay.2ka.mipt.ru ([194.85.82.65]:53749 "EHLO 2ka.mipt.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760859AbXHDRDw (ORCPT ); Sat, 4 Aug 2007 13:03:52 -0400 Date: Sat, 4 Aug 2007 21:03:44 +0400 From: Evgeniy Polyakov To: Manu Abraham Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org Subject: Re: Distributed storage. Message-ID: <20070804170344.GA18292@2ka.mipt.ru> References: <20070731171347.GA14267@2ka.mipt.ru> <1a297b360708022204u4fc7603pb6baebe2bdf28618@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1a297b360708022204u4fc7603pb6baebe2bdf28618@mail.gmail.com> User-Agent: Mutt/1.5.9i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1738 Lines: 35 On Fri, Aug 03, 2007 at 09:04:51AM +0400, Manu Abraham (abraham.manu@gmail.com) wrote: > On 7/31/07, Evgeniy Polyakov wrote: > > > TODO list currently includes following main items: > > * redundancy algorithm (drop me a request of your own, but it is highly > > unlikley that Reed-Solomon based will ever be used - it is too slow > > for distributed RAID, I consider WEAVER codes) > > > LDPC codes[1][2] have been replacing Turbo code[3] with regards to > communication links and we have been seeing that transition. (maybe > helpful, came to mind seeing the mention of Turbo code) Don't know how > weaver compares to LDPC, though found some comparisons [4][5] But > looking at fault tolerance figures, i guess Weaver is much better. > > [1] http://www.ldpc-codes.com/ > [2] http://portal.acm.org/citation.cfm?id=1240497 > [3] http://en.wikipedia.org/wiki/Turbo_code > [4] http://domino.research.ibm.com/library/cyberdig.nsf/papers/BD559022A190D41C85257212006CEC11/$File/rj10391.pdf > [5] http://hplabs.hp.com/personal/Jay_Wylie/publications/wylie_dsn2007.pdf LDPC codes require to solve N order matrix over finite field - exactly the reason I do not want to use Reed-Solomon codes even with optimized non-Vandermonde matrix. I will investigate LDPC further though. Turbo codes are like flow cipher compared to RS codes being block ciphers. Transport media is reliable in data storages, otherwise they would not even exist. -- Evgeniy Polyakov - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/