From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jakob Oestergaard Subject: Re: Is Read speed faster when 1 disk is failed on raid5 ? Date: Tue, 22 Oct 2002 12:45:22 +0200 Sender: linux-raid-owner@vger.kernel.org Message-ID: <20021022104522.GC24075@unthought.net> References: <02Oct22.043816edt.62658@gpu.utcc.utoronto.ca> 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: To: raid@ddx.a2000.nu Cc: Chris Siebenmann , linux-raid@vger.kernel.org List-Id: linux-raid.ids On Tue, Oct 22, 2002 at 11:58:56AM +0200, raid@ddx.a2000.nu wrote: > On Tue, 22 Oct 2002, Chris Siebenmann wrote: > > At a guess: when RAID-5 is intact, the kernel must read the entire > > stripe size worth of data to verify the parity. When a RAID-5 array= has > > a dead disk, any intact data block can be handed to the application > > the moment it comes off the disk as there is no point in doing pari= ty > > verification. (The kernel has to reconstruct the missing blocks, bu= t > > most of the blocks are present and good.) >=20 > i thought raid5 was doing parity calculation when writing ? Correct. RAID-5 will not do parity calculation when reading. RAID-5 wil= l give you availability but it will not give you integrity. > difference between rading from failed an intact raid5 is almost twice= the > speed (failed is doing 150MB/sec, intact is doing 80MB/sec) I didn't follow the beginning of this thread, sorry. Is this SCSI? --=20 =2E............................................................... : jakob@unthought.net : And I see the elder races, : :.........................: putrid forms of man : : Jakob =D8stergaard : See him rise and claim the earth, : : OZ9ABN : his downfall is at hand. : :.........................:............{Konkhra}...............: - 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