Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!simtel!news.kei.com!news.mathworks.com!tank.news.pipex.net!pipex!howland.reston.ans.net!Germany.EU.net!news.dfn.de!RRZ.Uni-Koeln.DE!se
From: se@MI.Uni-Koeln.DE (Stefan Esser)
Newsgroups: comp.periphs.scsi,comp.os.linux.hardware,comp.unix.bsd.freebsd.misc,comp.os.linux.setup
Subject: Re: SCSI PCI host adapter
Date: 4 Sep 1995 20:28:20 GMT
Organization: Institute for Mathematics, University of Cologne, Germany
Lines: 49
Message-ID: <42fnh4$egn@news.rrz.uni-koeln.de>
References: <418r3m$9c6@trauma.rn.com> <419963$hst@gate.sinica.edu.tw> <41eadg$1ajc@news.mindspring.com> <41rqk7$2oj@mailhub.hcl.com> <41sjek$rtu@bonnie.tcd-dresden.de> <FOO.95Aug31214507@yoda.fdt.net>
NNTP-Posting-Host: sysiphos.mi.uni-koeln.de
To: foo@yoda.fdt.net (R. M. Star)
Bcc: se
Xref: euryale.cc.adfa.oz.au comp.periphs.scsi:36566 comp.os.linux.hardware:14624 comp.unix.bsd.freebsd.misc:5338 comp.os.linux.setup:18753
In article <FOO.95Aug31214507@yoda.fdt.net>, foo@yoda.fdt.net (R. M. Star) writes:
|> In article <41sjek$rtu@bonnie.tcd-dresden.de> j@bonnie.heep.sax.de (J Wunsch) writes:
|> I remember some figures that claimed the AHA2940 could get about 20 %
|> more than the NCR.
|>
|> while the 2940 is a faster card, it's also much more expensive. an AHA2940
|> will cost around 190, whereas an NCR53c810 (biosless) is a mere 60 (i just
|> bought one)... and the 815 w/ bios is around 110. it makes much more sense
|> if you have multiple drives to buy may NCR's and put less drives on each,
|> allowing each drive a large portion of the scsi bus.
Well, but it's just an often heard misunderstanding,
that the NCR was slower than an Adaptec 2940 ...
From Russell Carters benchmark WWW pages:
-------Sequential Output-------- ---Sequential Input-- --Random--
-Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Disk MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
ST15150N1 100 3170 97.8 6430 55.0 2408 27.2 3395 97.6 6461 49.0 103.5 8.3
ST15150N2 100 3225 98.3 6199 46.6 2344 22.2 3441 96.8 6564 35.9 121.5 7.8
One of those is a P90 with 32MB of RAM and a NCR 53c810,
the other a P90 (or P100 ?) with 64MB of RAM and an AH2940.
(Well, you can see the effect of the larger RAM on the seek
times reported. I'd attribute the CPU load difference to the
different OS versions used; the AH2940 tests were done with
a very recent FreeBSD kernel, and I guess todays kernel might
result in even better values :)
For a second reference, I'm including the numbers posted by
Martin Cracauer <cracauer@wavehh.hanse.de> a few days ago:
-------Sequential Output-------- ---Sequential Input-- --Random--
-Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Disk MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
P100-1 140 1543 98.6 5150 40.3 1711 16.5 1808 98.6 5211 26.8 98.1 7.1
P100-2 140 1584 98.9 3138 20.5 1707 19.3 1850 98.8 5209 25.3 97.6 6.1
Now, which one is the NCR, which one the Adaptec ?
--
Stefan Esser, Zentrum fuer Paralleles Rechnen Tel: +49 221 4706021
Universitaet zu Koeln, Weyertal 80, 50931 Koeln FAX: +49 221 4705160
==============================================================================
http://www.zpr.uni-koeln.de/staff/esser/esser.html <se@ZPR.Uni-Koeln.DE>