From owner-freebsd-scsi Fri Jul 23 18:37:25 1999 Delivered-To: freebsd-scsi@freebsd.org Received: from pop.uniserve.com (pop.uniserve.com [204.244.156.3]) by hub.freebsd.org (Postfix) with SMTP id 27AC714DCA; Fri, 23 Jul 1999 18:37:20 -0700 (PDT) (envelope-from tom@uniserve.com) Received: from shell.uniserve.ca [204.244.186.218] by pop.uniserve.com with smtp (Exim 1.82 #4) id 117qjK-0000zH-00; Fri, 23 Jul 1999 18:36:02 -0700 Date: Fri, 23 Jul 1999 18:35:58 -0700 (PDT) From: Tom X-Sender: tom@shell.uniserve.ca To: David Gilbert Cc: Jeremy McMillan , mcmillan@cboe.com, freebsd-scsi@FreeBSD.ORG, freebsd-stable@FreeBSD.ORG Subject: Re: Postmark results with DPT RAID-5 In-Reply-To: <14232.62251.3136.872573@trooper.velocet.ca> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 23 Jul 1999, David Gilbert wrote: > >> I configured a DPT-3334UW on a K6-II/400 machine with 4 9G SCSI > >> drives. I found that for creating and deleting files, it couldn't > >> be beat (Ports would untar 4 times faster, for instance). However, > >> I found that a single 9G on a 2940UW would easily beat the raw > >> throughput. I was dissapointed with this. > > Jeremy> Why is this so? How's your DPT cache configured? > > I talked with the driver author and somewhat determined that my level > of performance was expected. When you think about it, the 3334 is a > 68040 --- which can run as fast as 50Mhz. Now, I'm sure that they > have other special hardware there, but the thruput bottleneck on the > 3334 is likely the 68040. Not too likely. The 68040 is a rather nice CPU. Especially since all RAID calculations are offloaded onto ASICs chips. The 68040 is basically just a SCSI command processor. > ... This would also probably be why later cards are based on RISC > processors. Not too likely. The RISC CPUs are just cheaper than using the next generation 680x0 series. DPT has pumped up their ASICs though. More likely you were using RAID5. RAID5 writing speed isn't so fast. > Dave. > > -- > ============================================================================ > |David Gilbert, Velocet Communications. | Two things can only be | > |Mail: dgilbert@velocet.net | equal if and only if they | > |http://www.velocet.net/~dgilbert | are precisely opposite. | > =========================================================GLO================ > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-stable" in the body of the message > > Tom To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message