Why you choose ns3 and not another network software

[rabahifa...@gmail.com]

hello ns3 users,

First of all I would like to thank all the developers of this group who really helped me during all my thesis years, and at their head Prof.Thomas R, Henderson co-founder and lead maintainer of the ns-3 network simulator for his advices, and his answers on all our needs.

Soon I will defend my Ph.D. thesis, and I used NS3 as a programming software during my Ph.D. thesis years. I worked in my Ph.D. thesis on " mobility in UAV networks", now I receive a question from the examiners of my thesis, where they want to know why I used NS3 and not another network simulation software.
I need a relevant scientific answer, to support my choice of NS3 in my publications.

My best regards

[igs...@gmail.com]

1) There is a myth about ns-2 based simulations among the researchers—people often say that journals will hesitate to accept a research paper which is based on out-dated ns-2 simulator. Even some reviewers of a thesis may ask the same question – why you used an outdated, unmaintained software for your research?
2) Even though Omnet++ is having excellent support for simulating UAV networks, it may have some scalability issue, if the simulated UAV network is little bit big and complex. (According to my own belief)  even though omnet ++ has rich functionalities for event tracing and trace analysis, it is little hard to get a customized output/graph from the bulk outputs of omnet++ simulation.
3) Often ns-3 has been proven to be a good open-source simulator in terms of scalability.  (According to my own belief) the event trace output generation and trace analysis part of ns-3 is more customizable - which is much important in a simulation based research. 
 

Personally, I love to do UAV simulations using ns-3 because of the availability of another 3D visualization tool (for ns-3) called NetSimulyzer. 

So, the animation outputs of ns-3+NetSimulyzer will be almost equal to that of Omnet++. So  ns-3+NetSimulyzer is a good choice for a research simulation.
Even though the visualization part of a simulation (NetAnim / NetSimulyzer outputs) are not major concerns of a scientific research,  those outputs will give good impression during a presentation or review meeting - it will make people easily understand what our simulation is doing.
You may compare the outputs of UAV Network Visualiztion on Omnet++ and ns-3 in the following links:
UAV Network visualization with Omnet++
UAV Network visualization with ns-3+NetSimulyzer

In case, you already used this 3D visualization tool in your research, then you may tell it as another reason for the selection of ns-3.

So, the only scientific reason one may say about the choice of ns-3 simulator is : scalability.
(and of course, simplicity)

Because,  Omnet++ has some obvious, attractive things for simulating a UAV network.

If anyone differs in my views, then please share your views here.

Charles Pandian.

[Charles Pandian]

Check the page no: 19 of the following article
http://jultika.oulu.fi/files/nbnfioulu-201502111072.pdf
Yes, it is said that ns-3 can hold 360 Million nodes in a simulation. so definitely it is much more scalable than any other simulator including most of the non-free simulators.

Charles Pandian,

--
Posting to this group should follow these guidelines https://www.nsnam.org/wiki/Ns-3-users-guidelines-for-posting
---
You received this message because you are subscribed to the Google Groups "ns-3-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ns-3-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ns-3-users/f1f32a5b-63bc-4057-995e-748d43d27933n%40googlegroups.com.