[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: ns performance



> Hi,
> 
> I want to use ns to simulate a network with about 5000 multicast nodes
> and about 10000 multicast messages. Can someone please tell me if such a
> large network can be simulated with ns and how long one simulation of
> this scenario approximatly runs (order of seconds, hours or days). I
> want to use a Pentium-150 Computer with Linux or Windows NT.

Can you explain more about the scenario you're trying to simulate, only 
the number of nodes and messages does not give enough information.
What information do you want to collect, and what behavior do you want to 
study. Do you plan on using detailed simulations and study transient 
behaviors and maybe queuing effects... or can you use the session 
simulator and lose the transient and queuing behaviors but scale better ?

How many senders are in your scenario, and how many receivers (are you 
going to assume that all nodes are members of the group)... how many 
multicast groups are you using ?!

sorry for all the questions, I should be giving you answers !! :-), but 
as Polly pointed out, there are lots of factors that decide the run time 
or memory of a simulation, especially that ns provides several levels of 
abstraction that trade-off some details for time/memory,

Regars,
-A

> 
> Thanks for your help,
> Christian
>