Control Plane Performance in Tactical Software Defined Networks

Abstract

Software Defined Networking can be beneficial in tactical networks to increase flexibility, provide programmability and to simplify management. The immense dynamics in tactical communication networks can, however, lead to excessive control traffic. As capacity in such networks is limited, this is a challenge that must be resolved. In this paper we investigate the control traffic overhead and present an analytical model that can predict the number of control messages for SDN networks with a given size and packet loss probability. We verify the soundness of the model using network emulations, and show that there is a good match between the analytical estimates of control traffic and real measurements. The tools provide some important insights that can be used to design and improve tactical SDN networks.