Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 2 Dec 1996 23:37:09 -0500
From:      "David S. Miller" <davem@jenolan.rutgers.edu>
To:        msmith@atrad.adelaide.edu.au
Cc:        dyson@freebsd.org, dennis@etinc.com, kpneal@pobox.com, hackers@freebsd.org
Subject:   Re: TCP/IP bandwidth bragging
Message-ID:  <199612030437.XAA18483@jenolan.caipgeneral>
In-Reply-To: <199612030323.NAA08269@genesis.atrad.adelaide.edu.au> (message from Michael Smith on Tue, 3 Dec 1996 13:53:37 %2B1030 (CST))

next in thread | previous in thread | raw e-mail | index | archive | help
   From: Michael Smith <msmith@atrad.adelaide.edu.au>
   Date: Tue, 3 Dec 1996 13:53:37 +1030 (CST)

   Well, where to start.  Look at the author of "lmbench", and
   consider the politics of his particular disease.

Personal attacks are not necessary, please stick to logical
arguments.

   Then consider the SPECmark fiasco.  If you can't see the picture
   clearly then, I'm happy to supply two-by-four correction.

I am not familiar with all of issues here, so I cannot comment.  But I
do agree that benchmarks are laundered in inconspicuous and
questionable ways, certainly.

   Who said anything about SGI?  If I sit down with an embedded
   processor system and any sort of network-alive RTOS, I can make
   your numbers look sick.  How'd you like a negative TCP latency?
   It's not hard to do; nor is 100% medium occupancy.

Are you offering full Unix (or POSIX, or some othe full featured
system) semantics on that system?  One of the things I am proud of is
that I can pretty much fill a nice pipe, and retain all of the
semantics of a full system.

   But that system's not going to run applications much better than
   Linux does, and this brings us back to John's point; micro-level
   benchmarks are for weiner-waving losers who fail to comprehend that
   computers are for _doing_ things with.

If you read John's other posting, he did not express his opinions on
this matter in this way at all.

---------------------------------------------////
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?199612030437.XAA18483>