*BSD News Article 73636


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.hawaii.edu!news.uoregon.edu!newsfeed.internetmci.com!in2.uu.net!ott.istar!istar.net!news.nstn.ca!newsflash.concordia.ca!newsfeed.pitt.edu!dsinc!news.acsu.buffalo.edu!news.drenet.dnd.ca!crc-news.doc.ca!nott!cunews!not-for-mail
From: shaver@neon.ingenia.com (Mike Shaver)
Newsgroups: comp.os.linux.networking,comp.unix.bsd.netbsd.misc,comp.unix.bsd.freebsd.misc
Subject: Re: TCP latency
Followup-To: comp.os.linux.networking,comp.unix.bsd.netbsd.misc,comp.unix.bsd.freebsd.misc
Date: 13 Jul 1996 20:10:18 GMT
Organization: Slacker Software Services
Lines: 84
Message-ID: <4s8vra$6ti@bertrand.ccs.carleton.ca>
References: <4paedl$4bm@engnews2.Eng.Sun.COM> <x7687w1dsr.fsf@oberon.di.fc.ul.pt> <4s220u$nmq@symiserver2.symantec.com> <31E53C2B.41C67EA6@inuxs.att.com> <4s6k8o$4o0@fox.ksu.ksu.edu> <31E6FD92.41C67EA6@dyson.iquest.net> <4s8cuq$ljd@bertrand.ccs.carleton.ca> <31E7C0DD.41C67EA6@dyson.iquest.net>
NNTP-Posting-Host: neon.ingenia.com
X-Newsreader: TIN [UNIX 1.3 950824BETA PL0]
Xref: euryale.cc.adfa.oz.au comp.os.linux.networking:45119 comp.unix.bsd.netbsd.misc:4028 comp.unix.bsd.freebsd.misc:23496

John S. Dyson (toor@dyson.iquest.net) wrote:
: Mike Shaver wrote:
: > John S. Dyson (toor@dyson.iquest.net) wrote:
: > : He made a claim that BSD networking is not as fast as Linux, specifically
: > : using the no-load latency figure.
: > 
: > He posted numbers to back up his claim that a given Linux was faster
: > than a given *BSD under a given load state on given hardware.
:
: Yes I am disputing the fact, the fact is that he had said that the
: TCP latency is faster.  Bzzt, that is the wrong conclusion.  The
: TCP latency under NO LOAD is faster.

Please re-read that.  It sounds suspiciously like we agree here. =)
We're talking about no-load latency.  That has _never_ been disputed.
If you want to say that the performance characteristics of the systems
are different under load, then you have been repeatedly (and
enthusiastically =) ) invited to produce some benchmarks that
demonstrate this.  I'm not saying that you're _wrong_, I'm saying that
you haven't given me any reason to believe you.

The only benchmark that's been presented here has demonstrated that,
under certain circumstances, Linux outperforms *BSD.  However narrow
and irrelevant you feel that benchmark is, it's still the _only_ set
of numbers we have.

(You might ask, quite reasonably, why you should give me that reason.
It appears to me from this discussion that you are _very_ interested
in having people believe you. =) )

: Most people don't understand
: the difference, but one claim is accurate, and the other is NOT.

Provably NOT?  (Not provably so, perhaps, but no-one is asserting that
it is.)

: The point is being missed, and this is why either Linus either doesn't
: know the scalability issues, or he is disinforming.  The issue is that
: when most people who use the OS see the latency problem, it is when the
: systems are under heavy load, and the latency that was measured by lat_tcp
: is pretty much meaningless.

: The only arrogance that I have seen is that certain people are trying
: to pass off benchmark results with bogus conclusions.

I refer you to your earlier post in which you dismiss lat_tcps
no-load latency measurements as something that most FreeBSD users
don't consider meaningful.

I've reread most of this thread, and can't find _anyone_ asserting
that the lat_tcp numbers are proof of anything other than no-load
latency.  I'd love a reference if you've got one.

: I am being
: skeptical because of the continued INCORRECT conclusions given the
: benchmark results presented.  (Psst, they are misinforming you and
: I am trying to protect you from that...)

Thanks.  Quis custodiet ipsos custodes?

: How can it POSSIBLY show ANYTHING but NO-LOAD latency?

It _can't_.  Luckily for the forces of Truth, Justice and the TCP Way,
no-one is assering otherwise.

: If a claim is made other than NO-LOAD latency, then someone is trying to
: pull the wool over your eyes...

Sure.  Great.
Was such a claim being made?  You were asserting that loaded latency
was more meaningful, but I think that's the only reference to loaded
latency I can think of.  Some Linux people probably think that our
loaded latency is better, too, but apparently some *BSD people think
the opposite.  Since there are no numbers being presented, we'll never
know.

Mike

-- 
#> Mike Shaver (shaver@ingenia.com) Ingenia Communications Corporation <#
#>       Chief System Architect -- Head geek -- System exorcist        <#
#>                                                                     <#
#>   "Have you considered a life?  I hear they're quite affordable     <#
#>          these days." --- shields@tembel.org                        <#