From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Thu, 22 Mar 2001 10:53:44 +0100 From: =?iso-8859-1?Q?Ragnar_Kj=F8rstad?= Subject: Re: [linux-lvm] Apparent performance degradation for each PV with striping Message-ID: <20010322105344.C11762@vestdata.no> References: <01C0B04F.5EFB3400.smdenton@bellsouth.net> <3AB60D44.3036CD9C@wrkhors.com> <0103220954110L.00851@lyta> Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable In-Reply-To: <0103220954110L.00851@lyta>; from Russell Coker on Thu, Mar 22, 2001 at 09:54:11AM +0100 Sender: linux-lvm-admin@sistina.com Errors-To: linux-lvm-admin@sistina.com Reply-To: linux-lvm@sistina.com List-Help: List-Post: List-Subscribe: , List-Unsubscribe: , List-Archive: List-Id: Content-Type: text/plain; charset="iso-8859-1" To: linux-lvm@sistina.com Cc: Steven Lembark On Thu, Mar 22, 2001 at 09:54:11AM +0100, Russell Coker wrote: > On Monday 19 March 2001 14:44, Steven Lembark wrote: > > Escalade is a nice idea, but still runs afoul of the interrupt > > & bus design of PC's. for high-i/o applications "PC" hardware > > doesn't work all that well. main problem here is delivering the > > raw datat to an Esc. controller through the existing PC. the > > add'l cpu load in this case comes from the bottom-half of drivers > > on heavily loaded cards that spend too much time waiting for > > access to the shared card. > > > > these are a distinct improvement over stock IDE or software RAID > > but don't expect them to suddenly turn your PC into a SparcServer > > or K400. >=20 > What is a K400? >=20 > Every test that I run shows SPARC machines running slower than desktop=20 > machines with IA32 CPUs. My Athlon-800 machine has more memory bandwidth= =20 > than an E450 according to the industry standard "streams" benchmark and=20 > according to a little memory benchmark I wrote. Also a single IBM 46G AT= A=20 > drive in my Athlon outperforms A1000 arrays in E450 class hardware. I second this. Every test we've run show that x86 linux boxes kick sparc but. :-) x86 with scsi-scsi RAID controllers we se performance 5-6 times that of E450s with A1000 RAID. Linux has a reputation for not matching other systems on NFS performance, but our tests show x86 linux beeing faster than both x86 freeBSD and SPARC Solaris (both client and server-side) However: the escalade RAID-controller doesn't have any write-back cache, does it? So performance actually be worse than on single disk systems (in some situations - see thread on reiserfs-list), and won't even compare to a real scsi-scsi or fc-scsi RAID-system. Anyway - we're getting off-topic awfully fast here; please send followups directly pr mail and remove lvm-list. --=20 Ragnar Kj=EF=BF=BDrstad Big Storage