A unified control and optimization framework for dynamical service chaining in software-defined NFV system

In software-defined radio and cognitive radio enabled future 5G mobile networks, and SDNNFV system will enable high flexibility in constructing service chaining, that is, steering flows through required service chains, thanks to the ability of dynamic service deployment and agile traffic routing. Given such flexibility, we need to carefully design the system to identify the optimal mechanism that maximizes both performance and resource utilization. To enable the SDN-NFV system to optimize the service chaining according to the user requirements and network environment, in this article, we propose a unified control and optimization framework that jointly controls and optimizes the resource allocation in networking and service provisioning. By modeling the network and services together and developing optimization techniques that jointly examine both, this framework benefits the overall system performance in various scenarios of traffic steering, VM selection, function assignment, and policy optimization. Furthermore, these achievable benefits are demonstrated by a typical example of simulation under a realistic scenario.