From mboxrd@z Thu Jan 1 00:00:00 1970 From: Keld =?iso-8859-1?Q?J=F8rn?= Simonsen Subject: Re: which raid level gives maximum overall speed? (raid-10,f2 vs. raid-0) Date: Tue, 5 Feb 2008 21:04:23 +0100 Message-ID: <20080205200423.GB17621@rap.rap.dk> References: <20080130192133.17b254bf@szpak> <20080130220007.GG20173@rap.rap.dk> <20080130233639.5d46b044@szpak> <20080131015506.GB6617@rap.rap.dk> <20080205161036.GA30326@rap.rap.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-raid-owner@vger.kernel.org To: Justin Piszcz Cc: Janek Kozicki , linux-raid@vger.kernel.org List-Id: linux-raid.ids On Tue, Feb 05, 2008 at 11:54:27AM -0500, Justin Piszcz wrote: >=20 >=20 > On Tue, 5 Feb 2008, Keld J=F8rn Simonsen wrote: >=20 > >On Thu, Jan 31, 2008 at 02:55:07AM +0100, Keld J=F8rn Simonsen wrote= : > >>On Wed, Jan 30, 2008 at 11:36:39PM +0100, Janek Kozicki wrote: > >>>Keld J=F8rn Simonsen said: (by the date of Wed, 30 Jan 2008 23= :00:07=20 > >>>+0100) > >>> > >> > >>All the raid10's will have double time for writing, and raid5 and r= aid6 > >>will also have double or triple writing times, given that you can d= o > >>striped writes on the raid0. > > > >For raid5 and raid6 I think this is even worse. My take is that for > >raid5 when you write something, you first read the chunk data involv= ed, > >then you read the parity data, then you xor-subtract the data to be > >changed, and you xor-add the new data, and then write the new data c= hunk > >and the new parity chunk. In total 2 reads and 2 writes. The read/wr= ites > >happen on the same chunks, so latency is minimized. But in essence i= t is > >still 4 IO operations, where it is only 2 writes on raid1/raid10, > >that is only half the speed for writing on raid5 compared to raid1/1= 0. > > > >On raid6 this amounts to 6 IO operations, resulting in 1/3 of the > >writing speed of raid1/10. > > > >I note in passing that there is no difference between xor-subtract a= nd > >xor-add. > > > >Also I assume that you can calculate the parities of both raid5 and > >raid6 given the old parities chunks and the old and new data chunk. > >If you have to calculate the new parities by reading all the compone= nt > >data chunks this is going to be really expensive, both in IO and CPU= =2E > >For a 10 drive raid5 this would involve reading 9 data chunks, and > >making writes 5 times as expensive as raid1/10. > > > >best regards > >keld > >- > >To unsubscribe from this list: send the line "unsubscribe linux-raid= " in > >the body of a message to majordomo@vger.kernel.org > >More majordomo info at http://vger.kernel.org/majordomo-info.html > > >=20 > On my benchmarks RAID5 gave the best overall speed with 10 raptors,=20 > although I did not play with the various offsets/etc as much as I hav= e=20 > tweaked the RAID5. Could you give some figures?=20 best regards keld - To unsubscribe from this list: send the line "unsubscribe linux-raid" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html