*BSD News Article 43696


Return to BSD News archive

Xref: sserve comp.unix.bsd.freebsd.misc:943 comp.unix.bsd.netbsd.misc:272 comp.protocols.tcp-ip:36229 comp.benchmarks:6962
Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.uwa.edu.au!classic.iinet.com.au!news.uoknor.edu!news.ecn.uoknor.edu!news.ysu.edu!usenet.ins.cwru.edu!howland.reston.ans.net!gatech!news.mathworks.com!zombie.ncsc.mil!simtel!col.hp.com!news.dtc.hp.com!hplextra!hplb!hpwin055.uksr!apollo.hp.com!hpax!cupnews2.cup.hp.com!raj
From: raj@cup.hp.com (Rick Jones)
Newsgroups: comp.unix.bsd.freebsd.misc,comp.unix.bsd.netbsd.misc,comp.protocols.tcp-ip,comp.benchmarks
Subject: Netperf should now work on *BSD and do TTCP (RFC 1644)
Followup-To: comp.benchmarks
Date: 5 May 1995 20:29:26 GMT
Organization: netperf
Lines: 25
Message-ID: <3oe1r6$fu2@hpindda.cup.hp.com>
Reply-To: raj@cup.hp.com
NNTP-Posting-Host: hpindio.cup.hp.com
Summary: Netperf 2.0PL1 for *BSD and TTCP test
X-Newsreader: TIN [version 1.2 PL2]

For those who would like to be benchmarking network performance on
[Free|Net]BSD platforms, netperf revision 2.0PL1 _should_ be working
without additional modification (although I've only run it on FreeBSD
myself). There are also some minor fixes for other platforms.

This version of netperf also includes a test for Transactional TCP
(RFC 1644) called TCP_TRR. To get the full force of TTCP, you need to
use the test-specific "-p portnum" option to cause the client side to
re-use the same local port number for each transaction. This test will
be compiled-in if you specify "-DDO_1644" in the CFLAGS of the
makefile.

Results can be compared with the TCP_CRR test, (which should behave
much like the http protocol) and perhaps the TCP_RR test. Before
drawing any general conclusions though, be sure to either get a full
explanation of the way each test works, or get one yourself by looking
at the source code.

You can find netperf, and the netperf results database at
http://www.cup.hp.com/netperf/NetperfPage.html

happy benchmarking,

rick jones
Please direct your follow-ups to comp.benchmarks.