Return to BSD News archive
#! rnews 1641 bsd
Newsgroups: comp.unix.bsd.bsdi.misc
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!nntp.coast.net!news.kei.com!newsfeed.internetmci.com!in1.uu.net!news.new-york.net!spcuna!spcvxb!terry
From: terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr.)
Subject: Re: Performance 3Com 3C509 vs. 3C590?
X-Nntp-Posting-Host: spcvxa.spc.edu
References: <4rkc94$q6h@ns2.mainstreet.net>
Sender: news@spcuna.spc.edu (Network News)
X-Nntp-Posting-User: TERRY
Organization: St. Peter's College, US
Date: Sun, 7 Jul 1996 07:54:55 GMT
Message-ID: <1996Jul7.035455.1@spcvxb.spc.edu>
Lines: 20
In article <4rkc94$q6h@ns2.mainstreet.net>, mwang@pandadesigns.com (Michael Wang) writes:
> I'm trying to spec a Pentium system for a heavy-traffic Web server and
> wanted to know if it was worth the few extra dollars to get the PCI
> version.
What speed is your link to the net? I'm amazed by people putting in 100Mbit
cards "to speed up access" via their T1 (1.536Mbit) links 8-)
OTOH, if you're seeing too much CPU time spend talking to a 3C509 (I'm
not sure how you'd tell, without instrumenting the kernel or implementing
the "interrupt time" stuff that's missing in BSD/OS) you might benefit from
a different card type. I'm trying DEC DE500's (10/100 PCI) as replacements
for my 3C509's, but I'll be running them in 10Mbit mode.
I expect that the transfer method (programmed I/O vs. DMA) matters a lot
more than the bus the cards are sitting on.
Terry Kennedy Operations Manager, Academic Computing
terry@spcvxa.spc.edu St. Peter's College, Jersey City, NJ USA
+1 201 915 9381 (voice) +1 201 435-3662 (FAX)