Re: Iperf for Wireless

To: iperf-users --at-- dast.nlanr.net
Subject: Re: Iperf for Wireless
From: Jon Dugan <jdugan --at-- ncsa.uiuc.edu>
Date: Tue, 16 Sep 2003 16:12:59 -0500
Content-disposition: inline
Content-type: text/plain; charset=us-ascii
In-reply-to: <EB55A4B03950DB4CBDAADD7D73D6EFC0042B79AF --at-- i2km32-ukdy.domain1.systemhost.net>
Reply-to: iperf-users --at-- dast.nlanr.net
Sender: owner-iperf-users --at-- dast.nlanr.net
User-agent: Mutt/1.4i

On Tue, Sep 16, 2003 at 05:34:48PM +0100, david.2.gould --at-- bt.com wrote:
> I am currently developing a wireless solution that involves throughput at
> different points in range.
> 
> I have tried using iperf, but with some trouble.  When I go out of range,
> iperf often reports throughput in UDP mode as succeeding, yet this is quite
> impossible!  Further the results in TCP mode are somewhat more odd.

  Where are you reading the results?  Since UDP is stateless the transmitter
  only knows how much data it sent, it has no way to determine how much was
  received.  Therefore if you are reading the results at the sending side it
  will look like no loss.  Iperf can communicate with the receiver and find
  out how much data the sender received, but this requires that there be
  reachability between the sender and the receiver. 

  TCP on the other hand provides reliability and thus will keep trying to
  resend data until it receives an acknowledgement from the receiver that the
  data has been received.

Jon
-- 
Jon Dugan             |  Senior Network Engineer, NCSA Network Research
jdugan --at-- ncsa.uiuc.edu  |  269 CAB, 605 E Springfield, Champaign, IL 61820
217-244-7715          |  http://www.ncsa.uiuc.edu/~jdugan/

References:
- Iperf for Wireless
  - From: david.2.gould --at-- bt.com

Prev by Date: Iperf for Wireless
Next by Date: TCP window size
Previous by thread: Iperf for Wireless
Next by thread: TCP window size

Other Mailing lists | Author Index | Date Index | Subject Index | Thread Index