Performance Evaluation of TCP/IP vs. OpenFlow in INET Framework Using OMNeT++, and Implementation of Intelligent Computational Model to Provide Autonomous Behaviour


Analysing the performance of transmitting data from a source to a certain destination is an interesting task. One of the most reliable networking protocol suites is the Transport Control Protocol and the Internet Protocol (TCP/IP), which will be studied against a new management paradigm called Software Defined Networking (SDN). SDN is an emerging programmable network architecture, where network control plane is decoupled from forwarding plane. The first standardize communication interface defined between the control and forwarding layers of SDN is known as OpenFlow. OpenFlow is a key enabler for SDN that allows direct manipulation of network devices through programmable applications that will fulfill the users’ desires. SDN forwarding methods are based on flows, which operate in contrast to conventional routing methods, such as TCP/IP routing table and MAC learning table. Moreover, OpenFlow protocol has efficient forwarding methods to push L2-L4 functions which are simplified into a Flow-Table(s) abstraction. This paper discusses the relationship between the processes of forwarding packets in conventional IP routing table vs. OpenFlow-table and evaluates the performance between both implementations using INET framework in OMNeT++. We argue that the placement of the controller can affect the performance of OpenFlow in terms of transporting data and conceding distances and costs. While OpenFlow performs slightly better than TCP with respect to mean round trip time (RTT). The results also proved the correctness of OpenFlow implemented simulation model. Finally, we propose the three phases of implementing a Distributed Active Information Model (DAIM) within OpenFlow to support an autonomic network management.

Author Information
Pakawat Pupatwibul, University of Technology Sydney, Australia
Ameen Banjar, University of Technology Sydney, Australia
Robin Braun, University of Technology Sydney, Australia

Paper Information
Conference: ACTIS2014
Stream: New Realities through Artificial Intelligence

This paper is part of the ACTIS2014 Conference Proceedings (View)
Full Paper
View / Download the full paper in a new tab/window

Posted by amp21