A comprehensive study of different objectives and solutions of controller placement problem in software‐defined networks
Mili Dhar, Arindam Debnath, Bidyut K. Bhattacharyya, Mrinal Kanti Debbarma, Swapan Debbarma
Abstract
Abstract Software‐defined network (SDN) is a programmable networking paradigm that enables logically centralized network management. In order to overcome the shortcomings of traditional networks, SDN has come into the picture. SDN is a recent and fast‐forwarding technique that involves separating the network's control plane from the data plane. This network also allows us to directly communicate with the applications through application programming interfaces. A network administrator can directly modify and update the routing policies and rules in SDN. Apart from the advantages given by the SDN, it has also been found that separation of control plane directly affects the network performance. Finding the optimum number of controllers and their locations to improve the performance is one of the fundamental research problems in SDN which is also commonly known as the controller placement problem (CPP). In this article, we have addressed the CPP in SDN in detail and investigated the optimization objectives that influence CPP. The state‐of‐the‐art of this article is to review almost all the optimization objective along with their mathematical formulation and survey the CPP solutions proposed by the other researchers'. Classified the CPP into three parts based on the objectives and solutions of CPP and reviewed the impact of these objectives on network performance. This article gives a comprehensive overview of current strategies on CPP in SDN and at last, we give some research issues that can be further explored by the researchers.