From mboxrd@z Thu Jan 1 00:00:00 1970 From: Justin Piszcz Subject: Re: Slow Soft-RAID 5 performance Date: Thu, 19 Jul 2007 12:03:46 -0400 (EDT) Message-ID: References: <469DE3BB.9070604@grupopie.com> <469F59E6.3050105@cgv.tugraz.at> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Return-path: In-Reply-To: <469F59E6.3050105@cgv.tugraz.at> Sender: linux-raid-owner@vger.kernel.org To: Lars Schimmer Cc: Linux RAID List-Id: linux-raid.ids On Thu, 19 Jul 2007, Lars Schimmer wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Justin Piszcz wrote: >> >> >> On Wed, 18 Jul 2007, Rui Santos wrote: >> >>> Hi, >>> >>> I'm getting a strange slow performance behavior on a recently installed >>> Server. Here are the details: >>> >>> Server: Asus AS-TS500-E4A >>> Board: Asus DSBV-D ( >>> http://uk.asus.com/products.aspx?l1=9&l2=39&l3=299&l4=0&model=1210&modelmenu=2 >>> >>> ) >>> Hard Drives: 3x Seagate ST3400620AS ( >>> http://www.seagate.com/ww/v/index.jsp?vgnextoid=8eff99f4fa74c010VgnVCM100000dd04090aRCRD&locale=en-US >>> >>> ) >>> I'm using the AHCI driver, although with ata_piix, the behavior is the >>> same. Here's some info about the AHCI controler: >>> >> >> With three disks, if everything was perfect yeah 120MB/s writes. When I >> had started out with 4 raptors I was getting 164MB/s read and write. By >> default with no optimizations you will not get good speed. >> >> With no optimizations with 10 raptors I get 180-200MB/s, with >> optimizations, 464MB/s write and 622MB/s read. >> >> 1. Use XFS if you wan't speed. >> 2. Use 128k, 256k or 1MiB chunk size. >> 3. Use 8192k, 16384k stripe_cache_size. >> 4. Use 65536 readahead size. >> >> These are only some of the optimizations I use. > > One question in between, which option of these I can set after RAID5 is > filled with data? > E.G. I setup a 9 disk SOFT-RAID5 with XFS and 3125669888 blocks level 5, > 128k chunk, algorithm 2 [9/9] [UUUUUUUUU]. > Speed is not really fast, I assume it could be faster. You can experiment with the stripe_cache_size and readahead size to start with. Justin.