*BSD News Article 64184


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.mel.connect.com.au!munnari.OZ.AU!news.ecn.uoknor.edu!news.cis.okstate.edu!news.ksu.ksu.edu!news.mid.net!newsfeeder.gi.net!newsfeed.internetmci.com!howland.reston.ans.net!blackbush.xlink.net!rz.uni-karlsruhe.de!news.uni-stuttgart.de!news.ruhr-uni-bochum.de!news.rhrz.uni-bonn.de!RRZ.Uni-Koeln.DE!zpr.uni-koeln.de!se
From: se@ZPR.Uni-Koeln.DE (Stefan Esser)
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: AMD 5x86 recommendations?
Date: 14 Mar 1996 19:37:55 GMT
Organization: Institute for Mathematics, University of Cologne, Germany
Lines: 57
Sender: se@Sisyphos (Stefan Esser)
Message-ID: <4i9sij$1b6@news.rrz.uni-koeln.de>
References: <rrwood-1103961728540001@bpci.net3.io.org> <Pine.LNX.3.91.960313193334.32655B-100000@gallup.cia-g.com> <slrn4kfn6l.95l.hwr@pilhuhn.de>
NNTP-Posting-Host: sisyphos.mi.uni-koeln.de
To: hwr@pilhuhn.de
Bcc: se

In article <slrn4kfn6l.95l.hwr@pilhuhn.de>, hwr@pilhuhn.de (Heiko W.Rupp) writes:
|> In <Pine.LNX.3.91.960313193334.32655B-100000@gallup.cia-g.com>, Stephen Fisher wrote:
|> :|(default drivers - test run) with an Adaptec 2940 SCSI controller and 16MB
|> :|of ram) and it took 2 hours and 31 minutes.  Sheesh, reminds me of my 
|> 
|> :|Doing the same (with IDE drives though..) on a 486DX4/100 with 16meg of 
|> :|ram took me about 30 minutes.
|> 
|> Yeh and a p90 under load with a bt946 does this in less then 10 minutes ..

As does an SP3G with AMD 5x86b and NCR PCI SCSI ...

|> Try changing your adaptec to a bt946 and retry the compile. I still
|> suspect the aha2940 to be very slow in FreeBSD.

No, it definitely isn't !

The AH2940 and NCR are nearly 
identical in performance.

Here my recent results with the 
AMD 5x86/NCR and a Quantum Atlas:

              -------Sequential Output-------- ---Sequential Input-- --Random--
              -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
Machine    MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU  /sec %CPU
5x86/Atls 100  2679 98.5  6027 54.9  2640 54.7  2846 97.9  6444 85.0  76.4  7.0

And now some Pentium results.
The following two lines are 
from Russell Carters web page
at Geli Engineering:

(http://www.geli.com/data/disk.perf.html)

              -------Sequential Output-------- ---Sequential Input-- --Random--
              -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
HP C3725N 100  5024 93.2  5134 23.6  1514 11.8  5085 82.6  5102 19.7  79.6  3.5
HP 3725C  100  5008 86.3  5065 22.8  1531 11.6  5121 82.2  5128 19.0  80.9  3.3

What do you think:

Which of these lines was 
obtained using the Adaptec,
which with the NCR 53c810 ?

Both used the same drive and 
identical hardware (a fast 
Pentium system; a P133 with 
PB cache, AFAIK).

Regards, STefan
-- 
 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/~se			  <se@ZPR.Uni-Koeln.DE>