From owner-freebsd-hackers Mon May 8 18:32:21 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id SAA28930 for hackers-outgoing; Mon, 8 May 1995 18:32:21 -0700 Received: from estienne.cs.berkeley.edu (estienne.CS.Berkeley.EDU [128.32.42.147]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id SAA28920 for ; Mon, 8 May 1995 18:32:18 -0700 Received: from localhost (localhost [127.0.0.1]) by estienne.cs.berkeley.edu (8.6.11/8.6.9) with SMTP id SAA03301; Mon, 8 May 1995 18:32:05 -0700 Message-Id: <199505090132.SAA03301@estienne.cs.berkeley.edu> X-Authentication-Warning: estienne.cs.berkeley.edu: Host localhost didn't use HELO protocol To: Tom Samplonius cc: Scott Mace , hackers@FreeBSD.org Subject: Re: Is a 486 fast enough for SCSI? In-reply-to: Your message of "Mon, 08 May 1995 15:34:11 PDT." Date: Mon, 08 May 1995 18:32:00 -0700 From: "Justin T. Gibbs" Sender: hackers-owner@FreeBSD.org Precedence: bulk > >On Mon, 8 May 1995, Scott Mace wrote: > >> > I've got a AMD486DX4100 with an Adaptec 2940 and two SCSI drives. When >> > I run iozone on one drive I get about 1.9MB/s, but when I run a iozone on >> > each drive I get about .9 MB/s (roughly half). Since the SCSI bus runs >> > at 10MB/s per second, the limiting factor appears to be the CPU? I >> > thought PCI devices required very little CPU time? >> > >> >> The limiting time is the speed of the drive. I can get over 3megs/sec >> with my EISA bt747 and seagate barracuda drives. And my cpu is only >> a 486DX2-66. I'm never seen a single scsi drive actually do 10MB/sec. > > I realize that no SCSI drive can do 10MB/s, but it is clearly not the >bottleneck when accessing two drives simultaneously, so what is? > >Tom I'm not sure since I don't have the same hardware. On my 2742 card, with a Quantum Empire 2100 and a Quantum PD1225 running "iozone 128 8192": Empire 2100 with the 1225 also doing an iozone: 3491843 bytes/second for writing the file 3106667 bytes/second for reading the file Emprire 2100 by itself. 4900133 bytes/second for writing the file 5204443 bytes/second for reading the file The first set of number looks a little low, but I'd have to play around with some of the bus timing numbers to get a better idea of just how bad they are. It isn't a 50% reduction though. -- Justin T. Gibbs ============================================== TCS Instructional Group - Programmer/Analyst 1 Cory | Po | Danube | Volga | Parker | Torus ==============================================