*BSD News Article 73741


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.ecn.uoknor.edu!news.eng.convex.com!newshost.convex.com!bcm.tmc.edu!pendragon!news.msfc.nasa.gov!newsfeed.internetmci.com!zdc!zdc-e!szdc-e!news
From: "John S. Dyson" <toor@dyson.iquest.net>
Newsgroups: comp.os.linux.networking,comp.unix.bsd.freebsd.misc
Subject: Re: TCP latency
Date: Mon, 15 Jul 1996 02:09:30 -0500
Organization: John S. Dyson's home machine
Lines: 40
Message-ID: <31E9EEAA.167EB0E7@dyson.iquest.net>
References: <4paedl$4bm@engnews2.eng.sun.com> <4pf7f9$bsf@white.twinsun.com> <4rql4p$39f@innocence.interface-business.de> <4rrimn$dro@fido.asd.sgi.com> <31E16DB5.41C67EA6@dyson.iquest.net> <4rtvpf$7e5@fido.asd.sgi.com> <31E73BCA.41C67EA6@FreeBSD.org> <4s7jsd$blf@fido.asd.sgi.com> <31E7B8A5.41C67EA6@dyson.iquest.net> <4s8rtp$jsh@fido.asd.sgi.com> <31E9B079.7A47@cet.co.jp>
NNTP-Posting-Host: dyson.iquest.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 3.0b5aGold (X11; I; FreeBSD 2.2-CURRENT i386)
Xref: euryale.cc.adfa.oz.au comp.os.linux.networking:45234 comp.unix.bsd.freebsd.misc:23587

Michael Hancock wrote:
> 
> Linux RULES!!!  Linux is the HEIR TO UNIX!!!!  Long Live LINUX!!!
> 
:-), silly isn't it...

Isn't it amazing, an OS that has rewritten the networking stack
to end up getting essentially the same performance as the BSD
stack (of course only demonstrated under no-load) and was developed
years later.  Sounds like they have been running in place :-).

Considering the effort that Linux has taken, re-working the
existing free OS would have been much more efficient.  I guess
the good thing is now that we have two major Free OSes.

It was so interesting to see the lmbench results posted by
a previous poster and how when the code was "done right"
in Linux, that it didn't perform significantly better/worse.
In fact there was a very curious difference in running a
real program (the shell results), as opposed the the
hello world test. I won't interpret the results, but the
numbers speak for themselves.  This is part of what I have
been talking about where small benchmarks don't
always show real-world performance.  Kind-of makes Linux look
like an educational exercise doesn't it?  If Microsoft could
sell MSDOS so long, I guess Linux can make it also. :-).

Since it is obvious that FreeBSD and Linux under light load
appear to run with roughly the same performance (plus or
minus some design tradeoffs), I suggest that Linus join
the FreeBSD team to continue on the long lived BSD development
effort.  This will allow us to start with our already
very good light AND heavy load performance and improve
it further (there is alot more scalability work yet to
do), and of course activate some of our more esoteric
features.  Wouldn't it be a much more efficient use
of resources?

John