[Ns-developers] ns-3 and future development

[mathieu lacage]

hi daniel,

Daniel Mahrenholz wrote:

> Before we spend to much effort in parallelization we should think 
> about for what we will use it. We had a master thesis that tried to 
> extend ns-2 to be useable for a parallel/distributed simulation of 
> MANETs. We


could you point me to a reference to this work ?

> learned that this is almost impossible to gain a speed-up using 
> conservative synchronization even using shared memory communication 
> (libSync). I don't know how PDNS performs for wired networks  or if 
> optimistic synchronization would  perform better but we should 
> investigate this before.


The people who worked on gtnets seem to think otherwise. They have 
published a few papers which show interesting results. Clearly, I have 
not attempted to reproduce them and to see how they apply to other 
simulation scenarios so, I cannot be positive this path will lead us to 
something useful.

> There is one thing that would be very useful (at least for me) - 
> parallelization/distribution of independent replications and 
> aggregation of results/traces in one place.


I see. Yes, it would be useful to you and other users (me included). 
However, I see this sort of problem as being decorelated from the 
simulator itself: it seems to me to be a matter of writing the right 
scripts to control the simulator. I wonder if such scripts could be made 
generic enough to be useful to more than one user.

regards,
Mathieu