Not much more than 10mbps?

Robert G. Brown rgb@phy.duke.edu
Wed Sep 30 17:46:43 1998


On Wed, 30 Sep 1998, Gerhard Sittig wrote:

> On Wed, 30 Sep 1998, Rogier Wolff wrote:
> > 
> > I have two computers connected by a crossover cable, and I don't see
> > the expected 100mbps performance.
> > 
> > [ ... ]
> > 
> > Yesterday I got ping delays of about 0.3 ms. Nice. But sometimes (say
> > about 20%) I got about 10ms or even 20ms. ftp would do about 1.2 to
> > 1.7 Mb per second. Even copying a raw disk (i.e. no seeks!) didn't
> > give the expected performance.
> 
> SHOULD these protocols show such a performance ?  AFAIR I never had

Yes, they should.  The right tool to use to test and verify is netperf.
With netperf I (and many others) have verified 97-99% data transmission
efficiencies (and 99%+ wirespeed efficiencies) on a switched connection,
and a crossover wire connection ought to be able to get this too -- it
is uniquely "switched".

To find (links to) netperf and a discussion of how to measure network
performance, see

www.phy.duke.edu/brahma/

and the (now rather old) network performance graphs thereupon.

   rgb

Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb@phy.duke.edu