Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 2 Dec 1996 23:19:32 -0500
From:      "David S. Miller" <davem@jenolan.rutgers.edu>
To:        jkh@time.cdrom.com
Cc:        dyson@FreeBSD.org, dennis@etinc.com, kpneal@pobox.com, hackers@FreeBSD.org, torvalds@cs.helsinki.fi, lm@engr.sgi.com, iain@sbs.de, sparclinux@vger.rutgers.edu
Subject:   Re: TCP/IP bandwidth bragging
Message-ID:  <199612030419.XAA18477@jenolan.caipgeneral>
In-Reply-To: <14423.849582120@time.cdrom.com> (jkh@time.cdrom.com)

next in thread | previous in thread | raw e-mail | index | archive | help
   Date: Mon, 02 Dec 1996 19:02:00 -0800
   From: "Jordan K. Hubbard" <jkh@time.cdrom.com>

   And how did Linus get dragged into this, anyway?

He got added for the same reasons the key FreeBSD developers are on
this CC: list.  Because I felt that linus could provide more
comprehensive technical arguments for some of the issues I have
brought up.

   Is this the modern equivalent of calling for Daddy, or what?

If I recall correctly, the BSD camps originated the proliferation of
entities such as "trolls" whose sole purpose is to just go "yeah,
yeah, what so and so said".

I'm not calling for Daddy, if you think this is so then you are
gravely informed.  Stop bringing up a straw man and shooting him down.
With that logic I could say that having Dyson on this CC: list is for
the same reason.  No one has commented to such and end.

   > lmbench numbers for all purchases these days?  What concrete
   > numbers are you able to put on that tally sheet?  None, because
   > whatever benchmarks the freebsd people are using to perform their
   > improvements are under lock and key, most likely because once the
   > Linux crowd had these at their disposal, we'd fix the problems
   > they show because

   You're gravely misinformed.

I don't feel that I am.  Please show me that it is untrue that raw
lmbench numbers do not translate into sales for anyone, and I'd be
equally interested in seeing claims to the contrary that the governemt
specifically specs lmbench numbers.  I would not make such a claim if
I were not certain of it.

   First off, you assume that we play the numbers game in the same way
   that you do when we manifestly don't.  We don't brag out about our
   lmbench numbers just as we don't walk around with our pants off
   carrying rulers and comparing penis lengths, and I think this whole
   silly thread all started off in reaction to your .signature.  If
   public a display of your bulging manhood is what it takes to floats
   your boat then have at it, but don't expect us to play the same
   game and certainly don't castigate us for refusing to play.

Someone made the choice to comment loudly about my .sig, they could
have just as well ignored it and not CC:'d their comments to me on top
of it.

   Second, we don't have a "tally sheet" because we know that a tally
   sheet *for the numbers we find meaningful* would be essentially
   useless for anyone else's comparison purposes.

Are you saying it wouldn't have the effect of selling systems just
like I am claiming lmbench numbers do?

   Given the wide disparity between what we feel to be important and
   what you feel to be important, how is any meaningful comparison
   even possible?  Take the hint - it's not, at least not until/unless
   you build some truly beefy servers we can see numbers for.

I've stated beefy servers that run day and night and perform very
well.  Perhaps it is easy for you to drop those things which I have
said.

   I hope for the sake of your tests that you increase this well
   beyond 50 work stations.  Get someone to put the server at a NAP
   and (I'm perfectly serious) find some free porn or popular
   shareware to stick on it for a week.  That will attract more than
   enough web users, far more than 50, and you can see how the machine
   truly performs under the onslaught.  I'm sure there's got to be a
   Linux fan somewhere with a T3 or better connection who'd be willing
   to make this a meaningful test.

   I'd be interested in the results.  Until then, I think it's in
   everyone's best interest that this thread end.  We all have better
   things we should be doing.

I'd be interested as well.

---------------------------------------------////
Yow! 11.26 MB/s remote host TCP bandwidth & ////
199 usec remote TCP latency over 100Mb/s   ////
ethernet.  Beat that!                     ////
-----------------------------------------////__________  o
David S. Miller, davem@caip.rutgers.edu /_____________/ / // /_/ ><



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199612030419.XAA18477>