Re: [ns] interpreting a trace file

[Lloyd Wood <l.wood@eim.surrey.ac.uk>]

On 4 Dec 2001, ruhiya mahalati wrote:

> i am beginner to NS and am working on a simulation project. I need to calculate average packet delay. I guess this can be done from the trace file. Can someone please tell me how to interpret a tracefile. I am using exponential distribution and gamma distribution. my trace for exponential distribution looks something like this :
>
> + -t 56.6600092128397 -s 3 -d 5 -p exp -e 1000 -c 0 -i 1 -a 0 -x {3.0 7.0 0 ------- null}
>
> - -t 56.6600092128397 -s 3 -d 5 -p exp -e 1000 -c 0 -i 1 -a 0 -x {3.0 7.0 0 ------- null}
>
> h -t 56.6600092128397 -s 3 -d 5 -p exp -e 1000 -c 0 -i 1 -a 0 -x {3.0 7.0 -1 ------- null}
>
> + -t 56.6600092128397 -s 3 -d 5 -p exp -e 440 -c 0 -i 2 -a 0 -x {3.0 7.0 1 ------- null}
>
> - -t 56.7000092128397 -s 3 -d 5 -p exp -e 440 -c 0 -i 2 -a 0 -x {3.0 7.0 1 ------- null}
>
> h -t 56.7000092128397 -s 3 -d 5 -p exp -e 440 -c 0 -i 2 -a 0 -x {3.0 7.0 -1 ------- null}
>
>
> how do i find out which columns stand for timestamp?

You could read the manual - but the slowly-increasing floating-point
value is a safe bet, no?

I wrote an awk script (on my ns webpage) to parse the 'normal'
tracefile to compute packet delays; it could be easily changed to use
this format.

L.

<L.Wood@surrey.ac.uk>PGP<http://www.ee.surrey.ac.uk/Personal/L.Wood/>