Date: Mon, 12 Feb 2001 02:22:02 -0800 From: "Ted Mittelstaedt" <tedm@toybox.placo.com> To: "Mike Meyer" <mwm@mired.org> Cc: <questions@freebsd.org> Subject: RE: Problems installing 4.x on large disks Message-ID: <002c01c094dd$a3c8c120$1401a8c0@tedm.placo.com> In-Reply-To: <14983.36943.315670.474001@guru.mired.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> -----Original Message----- > From: Mike Meyer [mailto:mwm@mired.org] > Sent: Sunday, February 11, 2001 11:27 PM > To: Ted Mittelstaedt > Cc: Mike Meyer; questions@freebsd.org > Subject: RE: Problems installing 4.x on large disks > > > > And I'll be the first to say that this comparison is pretty much > meaningless. The two systems we did this on were *totally* different, > and all the advantages are in the IDE drive favor (except CCD, of > course). His system is running FreeBSD 2.2.8 vs. 3.x (probably 3.3) on > mine; I had freshly formatted file systems, and he was working on a > news spool. I suspect I've got a faster CPU (dual PII/Xeons 400s) and > possibly system bus (100MHz) than his test system did as well, because > a news server doesn't need that kind of horsepower. > Test file size makes a tremendous difference. I don't know how much ram you had in the system but the docs for bonnie say you need to do 2X the amount of system ram for the sample file. In my case the system is a Pentium 200 and the amount of ram is 128MB. Also, the system bus was the 33Mhz stock PCI. But the CPU speed is actually what most likely accounts for the biggest difference. I did try bonnie on the news spool while the server was running, with a 1MB sample file just to see what the representation of writing a single article would be like and the times were very similar to yours. Of course I'm assuming that the bonnie test file probably spent most of it's time in the disk cache. > > Well, once you take that "bland PC" and install FreeBSD on it, you've > got a mouse, a GUI, Unix, and a complete software development > environment. Sounds like a Unix workstation to me. It'll also > outperform most of the RISC workstation I've dealt with, even though > they had SCSI. I think the RISC vs CISC argument was put to bed a while ago. :-) > What more do you want? Reliability immediately springs to mind, although a tapedrive for backup would be a useful peripheral to start with. These days the reliability component is highly dependent on the motherboard/cpu/processor selection, and in even the last 12 months there has been almost a revolution in motherboard manufacturing. The Taiwanese motherboard manufacturers have finally seem to have gotten it right, or perhaps competition due to the Athlon has finally put the fear of lost markets into Intel and AMD and as a result they are working closely with the chipset and board designers. > > You say "cost-is-no object workstation", well that's silly, what do you > > think > > that a "workstation" is? Last I checked, "cost-is-no-object" was an > > integral > > part of the definition of the word "workstation" > > It's certainly never been anywhere I've been. Most of them were > business trying to make a profit, and cost was *always* an > object. That's why most Unix workstations (and here I'm talking about > RISC boxes running commercial Unix) I've dealt with only had one disk > instead of striped disks, and so on. > Your probably only seeing the small companies then. It's a different world in the large ones. In the 2 large software firms I've worked for it was always the same - whenever an engineer needed a new workstation, you got the best, with total disregard of the cost. The reason for this was pretty simple, and logical - spending an extra 2-3K on hardware guarenteed that the hardware that you got was rock-solid reliable, because with hardware that was less reliable a possibility existed that a crash could wreck a day or 2 of production for a developer, and at the rates those guys were being paid, their lost time would eat up any $500-per-box savings that you could get. In the smaller firms under 200 employees I've worked for it's different - and this is the kind of thing that is the biggest problem that smaller firms have to overcome. The engineers and administrators cannot resist the lure of wasting time shopping for the best deal, and so getting a new workstation becomes a monthlong production for them. Instead of coding they waste hours of time looking for the fastest, cheapest and latest toys. So they then end up saving a grand on a PC but they do it by spending $5K in employee time and lost company profits. And the smaller the firm the worse it is, the longer they take to make a decision to buy something. > If you want to make that your definition of workstation, then I won't > argue with what you said. Of course that makes your objection pretty > much meaningless to almost everyone running FreeBSD, because they do > worry about cost. > If they are coming from a small company background yes they do - but of course their worrying is screwy, because instead of spending their production time worrying about this sort of thing, they would be better off going to Apache Digital or ASA Computers or someone like that and spending 5 minutes to order one of the $5K developers workstations, (note they are SCSI-based) and then getting back to work. Now, note that _I_ don't do that for my _own_ systems, but then I'm not paid for anything that I use them for. Ted Mittelstaedt tedm@toybox.placo.com Author of: The FreeBSD Corporate Networker's Guide Book website: http://www.freebsd-corp-net-guide.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?002c01c094dd$a3c8c120$1401a8c0>