Software-Defined Network (SDN) has emerged as an innovation framework with a separated control and data layer running upon Openflow protocol, and simplifies the deployment of the network. Though Software-Defined Network surely solves the specific problems that exist in traditional network, the evaluation on it is necessary. Currently, SDN-enable devices are rare and expensive. Existing simulators show a good performance in several aspects: realism, fidelity, arbitrary topology designing, and tools for debugging. So using software to simulate, monitor, and debug SDN is a relatively better and more economical approach. But the limitations on software are also noticeable when the focus shifts to a large-scale network designing, networks with multiple controllers, load simulation, and flexibility of testbed. Apart from those limitations, the debug information is still far from details expected. All those confine SDN experiments into a tiny space. To solve problems above, in this paper, we propose a scalable testbed framework for SDN simulations with multiple controllers, namely SFMC. Different from existing simulators, SFMC encompasses six modules. It ameliorates many functions as providing a clear structure, supporting several link types (including wireless link), extending the topology scalability into infinitive, enabling multiple controllers, and combining virtual and physical devices. We verify the performance of this testbed and show our experimental results from different scenarios, such as the flow simulation, WiFi environment, and multiple controllers.
[1]
Fang Hao,et al.
Towards an elastic distributed SDN controller
,
2013,
HotSDN '13.
[2]
Alex C. Snoeren,et al.
High-fidelity switch models for software-defined network emulation
,
2013,
HotSDN '13.
[3]
Stefan Schmid,et al.
Exploiting locality in distributed SDN control
,
2013,
HotSDN '13.
[4]
Paul Barford,et al.
Fast, accurate simulation for SDN prototyping
,
2013,
HotSDN '13.
[5]
Min Zhu,et al.
B4: experience with a globally-deployed software defined wan
,
2013,
SIGCOMM.
[6]
Basil S. Maglaris,et al.
Combining OpenFlow and sFlow for an effective and scalable anomaly detection and mitigation mechanism on SDN environments
,
2014,
Comput. Networks.
[7]
Julie A. McCann,et al.
UbiFlow: Mobility management in urban-scale software defined IoT
,
2015,
2015 IEEE Conference on Computer Communications (INFOCOM).