Re: benchmark terminology

Article View: comp.dcom.lans.fddi

Article #37

Back to comp.dcom.lans.fddi Previous Next

#37

From: Vernon Schryver
Date: Tue, 04 Aug 1992 15:24

24 lines
1493 bytes

In article <1992Aug4.124128.8089@alliant.com>, werme@alliant.com (Ric Werme) writes:
> ...
> While I have the floor here, may I suggest another (hah!) solution the
> the performance reporting confusion?  Suppose we reported throughput
> in terms of percentage of the maximum data rate?  We'd even have a
> chance of enforcing milli% to be 10^-3!

That might be reasonable.  It would cause some stress, because
determining and then agreeing on the maximum data rate is not easy.

For example, at InterOP91 there were reports of "100 Mbit/sec TCP over
FDDI."  Taken literally, that is impossible, because tokens, interframe
gaps, and so on consume a significant part of the potential FDDI
bandwidth.

The diameter of an FDDI ring, the value of TTRT, the presense of
"purger frames", higher layer protocol headers, and so on affect the
maximum possible FDDI bandwidth.  For example, a token latency of 50
milliseconds on a 50 microsecond ring seems to be good only for about
11.8 Mbit/sec of UDP/IP from a single station.

Vernon Schryver, vjs@sgi.com

Technical Details

Message-ID: <o5cuie4@rhyolite.wpd.sgi.com>

Path: arc1-ca.pugleaf.net!sparky!uunet!sun-barr!cs.utexas.edu!sdd.hp.com!mips!odin!sgi!rhyolite!vjs

References: <811@ceco.ceco.com> <1992Aug4.124128.8089@alliant.com>

Show Full Headers

Complete Article Headers

Newsgroups: comp.dcom.lans.fddi
Path: arc1-ca.pugleaf.net!sparky!uunet!sun-barr!cs.utexas.edu!sdd.hp.com!mips!odin!sgi!rhyolite!vjs
From: vjs@rhyolite.wpd.sgi.com (Vernon Schryver)
Subject: Re: benchmark terminology
Message-ID: <o5cuie4@rhyolite.wpd.sgi.com>
Organization: Silicon Graphics, Inc.  Mountain View, CA
References: <811@ceco.ceco.com> <1992Aug4.124128.8089@alliant.com>
Date: Tue, 4 Aug 1992 15:24:19 GMT
Lines: 24

Reply to Parent Message