[ndnSIM] sample-topologies

[Alex Afanasyev]

Hi Andriana,

The project that you're referring provides only partial information that can be used in simulation scenarios.  In particular, there were no delays and link bandwidths.  What we basically did is defined a basic heuristic to classify nodes into different classes (clients, gateways, and backbones---these what colors represent) and randomly assigned delays and bandwidths based on the classification.

Again, all these processing has no relation to the original Rocketfuel project, but provides some topology that can be used in the simulation, that has certain real-world characteristics (though not exact).  Depending on the simulation scenario you're working on, you may need to change heuristic or try to use some other topology...

For the definition of the heuristic, check the https://github.com/cawka/ndnSIM-sample-topologies/blob/master/scenarios/rocketfuel-maps-cch-to-annotaded.cc  source.

---
Alex

On Mar 20, 2014, at 5:54 AM, ioannoa at scss.tcd.ie wrote:

> Hello everyone,
> 
> I was currently looking at the sample topologies from Rocketfuel included
> into the same named directory for real-topology simulations.
> 
> However, I was not able to link the sample topologies with the known paper
> of:
> " Measuring ISP topologies with Rocketfuel". The ASes seem to have
> different number of nodes, either concerning routers or PoPs.
> 
> Does anyone have an idea of how these topologies have been concluded in
> ndnsim and what exactly each node, i.e. blue, green, red represents? My
> assumption is that blue nodes are backbone nodes, green nodes are gateway
> nodes and red nodes are local Points of Presence where clients are
> connected to.
> 
> Please, keep me informed if you have any ideas on it.
> 
> Kind regards,
> Andriana.
> 
> _______________________________________________
> ndnSIM mailing list
> ndnSIM at lists.cs.ucla.edu
> http://www.lists.cs.ucla.edu/mailman/listinfo/ndnsim