[ns] Confused about NS-2 capabilities

Martin McKenny mkenny@mekb2.sps.mot.com
Tue Apr 23 05:21:33 2002


--------------059606D0A4ABAD6D0FFE14E4
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: base64

Hi All,

I have recently changed my approach to simulating a new protocol.
Previously, I was sending packets one at a time over each link. However,
this proved to be insufficient for my requirements since I must be able
to interrupt a packet at given boundaries in order to send control
symbols such as acknowledge symbols.

Thus I altered my approach such that I now send packets using multiple
send() calls which send the packet one 'granule' at a time, where a
granule happens to be 4 bytes. Thus I invoke a for-loop that sends the
necessary number of granules to represent the packet. This works fine
until I have more than 12 or 13 granules representing a packet. After
this limit granules disappear and this completely destroys the
simulation behaviour.

Does anyone know of a limit when calling the send() method or the number
of packets that can be in transit over a link? Maybe I need to introduce
a delay between sending the
granules? Any help / suggestions / directions to relevant documentation
sections  appreciated.

Thanks,
Martin.

--
==============================================================
Martin McKenny
Research Engineer ISLI/Motorola
Phone  : +44 1506473378
Email  : mkenny@mekb2.sps.mot.com             (Mot)
       : martin.mckenny@sli-institute.ac.uk   (ISLI)
==============================================================



--------------059606D0A4ABAD6D0FFE14E4
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit

<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
Hi All,
<p>I have recently changed my approach to simulating a new protocol. Previously,
I was sending packets one at a time over each link. However, this proved
to be insufficient for my requirements since I must be able to interrupt
a packet at given boundaries in order to send control symbols such as acknowledge
symbols.
<p>Thus I altered my approach such that I now send packets using multiple
send() calls which send the packet one 'granule' at a time, where a granule
happens to be 4 bytes. Thus I invoke a for-loop that sends the necessary
number of granules to represent the packet. This works fine until I have
more than 12 or 13 granules representing a packet. After this limit granules
disappear and this completely destroys the simulation behaviour.
<p>Does anyone know of a limit when calling the send() method or the number
of packets that can be in transit over a link? Maybe I need to introduce
a delay between sending the
<br>granules? Any help / suggestions / directions to relevant documentation
sections&nbsp; appreciated.
<p>Thanks,
<br>Martin.
<pre>--&nbsp;
==============================================================
Martin McKenny
Research Engineer ISLI/Motorola
Phone&nbsp; : +44 1506473378
Email&nbsp; : mkenny@mekb2.sps.mot.com&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; (Mot)
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; : martin.mckenny@sli-institute.ac.uk&nbsp;&nbsp; (ISLI)
==============================================================</pre>
&nbsp;</html>

--------------059606D0A4ABAD6D0FFE14E4--