Modelling Software-Defined Networking: Software and hardware switches

Abstract In Software-Defined Networking (SDN), a switch is a forwarding device that moves data packets in a network. A software switch handles forwarding functions in software and thus cannot forward packet at line speed while a hardware switch leverages optimised forwarding hardware to forward packets at line speed. However, there has been very little research in the literature to help network engineers understand the tradeoffs in choosing one over the other. In this paper, we develop a unified queueing model for characterizing the performance of hardware switches and software switches in SDN. The unified queueing model is an analytical tool for engineers to predict delay, packet loss and throughput in their SDN deployments. Existing queueing models of SDN have focused on performance analysis of software switches, while our work presented herein is the first to present a unified analysis of hardware and software switches. Our proposed models exhibit errors below 5% compared to simulation. Between a hardware and software switch, the evaluation shows that a hardware switch achieves an average 80% lower delay and up to 100% lower packet loss probability compared to a software switch. The more a hardware switch involves the controller for decisioning, the lower the gains in terms of packet delays through the switch.

[1]  Azeem Iqbal,et al.  A stochastic model for transit latency in OpenFlow SDNs , 2017, Comput. Networks.

[2]  Ali Fanian,et al.  An analytical model for delay bound of OpenFlow based SDN using network calculus , 2017, J. Netw. Comput. Appl..

[3]  Simon Oechsner,et al.  Modeling and performance evaluation of an OpenFlow architecture , 2011, 2011 23rd International Teletraffic Congress (ITC).

[4]  Olav N. Østerbø,et al.  On The Modeling of OpenFlow-based SDNs: The Single Node Case , 2014, ArXiv.

[5]  Bryan Ng,et al.  Modelling Software-Defined Networking: Switch Design with Finite Buffer and Priority Queueing , 2017, 2017 IEEE 42nd Conference on Local Computer Networks (LCN).

[6]  Tseng-Chang Yen,et al.  An SDN-based cloud computing architecture and its mathematical model , 2014, 2014 International Conference on Information Science, Electronics and Electrical Engineering.

[7]  P. Burke.  The Output of a Queuing System , 1956 .

[8]  Hideaki Takagi,et al.  Stochastic Analysis of Computer and Communication Systems , 1990 .

[9]  Rakesh Kumar,et al.  End-to-End Network Delay Guarantees for Real-Time Systems Using SDN , 2017, 2017 IEEE Real-Time Systems Symposium (RTSS).

[10]  Maciej Kuźniar,et al.  What You Need to Know About SDN Flow Tables , 2015, PAM.

[11]  Bryan Ng,et al.  Queueing analysis of software defined network with realistic OpenFlow-based switch model , 2019, Comput. Networks.

[12]  Wei Li,et al.  Performance evaluation of OpenFlow-based software-defined networks based on queueing model , 2016, Comput. Networks.

[13]  P.E. Heegaard.  Evolution of Traffic Patterns in Telecommunication Systems , 2007, 2007 Second International Conference on Communications and Networking in China.

[14]  Ramin Yahyapour,et al.  An analytical model for software defined networking: A network calculus-based approach , 2013, 2013 IEEE Global Communications Conference (GLOBECOM).

[15]  Geyong Min,et al.  Performance Modelling of Preemption-Based Packet Scheduling for Data Plane in Software Defined Networks , 2015, 2015 IEEE International Conference on Smart City/SocialCom/SustainCom (SmartCity).

[16]  Alex C. Snoeren,et al.  High-fidelity switch models for software-defined network emulation , 2013, HotSDN '13.

[17]  Paul Goransson,et al.  Software Defined Networks: A Comprehensive Approach , 2014 .

[18]  Frank Dürr,et al.  Time-sensitive Software-defined Network (TSSDN) for Real-time Applications , 2016, RTNS.

[19]  Shui Yu,et al.  Modeling and performance analysis for multimedia data flows scheduling in software defined networks , 2017, J. Netw. Comput. Appl..

[20]  Thierry Turletti,et al.  A Survey of Software-Defined Networking: Past, Present, and Future of Programmable Networks , 2014, IEEE Communications Surveys & Tutorials.

[21]  Danda B. Rawat,et al.  Software Defined Networking Architecture, Security and Energy Efficiency: A Survey , 2017, IEEE Communications Surveys & Tutorials.

[22]  Sudip Misra,et al.  Buffer Size Evaluation of OpenFlow Systems in Software-Defined Networks , 2019, IEEE Systems Journal.

[23]  Geyong Min,et al.  Performance Modelling and Analysis of Software-Defined Networking under Bursty Multimedia Traffic , 2016, ACM Trans. Multim. Comput. Commun. Appl..

[24]  Marcel F. Neuts,et al.  Matrix-geometric solutions in stochastic models - an algorithmic approach , 1982 .

[25]  Ana Busic,et al.  Perfect Sampling of Networks with Finite and Infinite Capacity Queues , 2012, ASMTA.

[26]  Yong Xiang,et al.  Performance Analysis of Software-Defined Network Switch Using $M/Geo/1$ Model , 2016, IEEE Communications Letters.

[27]  Bernhard Plattner,et al.  Towards SDN based queuing delay estimation , 2016, China Communications.

[28]  Junjie Liu,et al.  The FlowAdapter: enable flexible multi-table processing on legacy hardware , 2013, HotSDN '13.

[29]  Cunqing Hua,et al.  Traffic-load aware user association in dense unsaturated wireless networks , 2014, 2014 Sixth International Conference on Wireless Communications and Signal Processing (WCSP).