Date: Mon, 1 Apr 1996 17:52:36 -0600 From: dkelly@hiwaay.net (David Kelly) To: Luigi Rizzo <luigi@labinfo.iet.unipi.it>, davidg@root.com Cc: dutchman@spase.nl, freebsd-hackers@freebsd.org Subject: Re: HDD cpu usage (IDE vs. SCSI). Message-ID: <v02140b03ad861446c295@[206.104.23.132]>
next in thread | raw e-mail | index | archive | help
At 9:45 AM 4/1/96, Luigi Rizzo wrote:
>> >How come the seagate uses twice as much cpu as the quantum?
>>
>> Because a Pentium-90 is much faster than a 486-100 for certain things and
>> most of the %CPU is for total I/O overhead, not just the overhead in the
>> device driver.
>
>Hmmm... if I get it right this means that under certain circumstances
>(1 disk, onboard IDE controller, medium-fast CPU) using SCSI instead
>of IDE gives you only a very little saving (which BTW is what I am
>convinced of, but this has not been the dominating opinion on this
>list).
No, that's not it at all. Standard Scientific Method: Only change one
variable at a time. Either put the SCSI card in the Pentium or the IDE
drives in the DX4/100. Repeat your test. Even then it won't be completely
accurate without nearly identical drives, same manufacturer, same model,
one SCSI, one IDE.
For real interesting information run both of your tests at the same time
(on the same CPU), one on SCSI, the other on IDE. Or two tests to two SCSI
drives on the same bus. Or two IDE drives on the same bus. Then you will
see the difference.
IDE does a perfectly good job of what it was designed for: cheap single
user single tasks. And that's what you were measuring. But by changing the
CPU's you also altered the units of your measurement (% of what? Answer "%
of Pentium, % of DX4/100, units don't match.").
--
David Kelly N4HHE, n4hhe@amsat.org, dkelly@hiwaay.net
=============================================================
To invent, you need a good imagination and a pile of junk.
- Thomas Edison
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?v02140b03ad861446c295>