RE: [ns] Strange DiffServ Results.

["Sergio Andreozzi" <sergio.andreozzi@lut.fi>]

> In the sample script, 2 UDP of 4M rate are used for both classes.
> therefore, I expected no change in the other class.
I suggest you an experimental work. Just try to put two counters in the WRR
scheduler. With these counters you want to count the number of times each
queue si found in idle status from the scheduler.

Check before and after the queue lenght modification. And if you find (as I
suppose) more idle status for the queue you "cut", you can say the other one
receive more service and so the number of dropped packet due to overflow
decrease.

Regards,


Sergio