[ns] limitations of ns for wireless, ad-hoc networks?

[Aggelos Bletsas <aggelos@media.mit.edu>]

Hi,

I'm a new user in ns and I would like to know the limitations of ns, for
the simulation of wireless, ad-hoc networks with a large number of
nodes.

It is apparent that for a number of 50 nodes and a simulated 900 secs
time, ns works consuming approximattely 4MB of memory per node producing
large traces (with the size on the order of hundreds of MBs depending on
what it is traced)

Apart from the requirements in time and disk space, are simulations of
let's say thousands of nodes for wireless networks feasible in ns?
If there is no problem, is there any estimate of the requirements in
memory needed per node? or the overall requirements of the hardware
needed to run those simulations?


thank you for your time,

Aggelos Bletsas