Dynamic Handler Framework for Network Slices Management

Flexibility, scalability and programmability are the main goals of future network design. To meet these goals, the network has to ensure a complete management and control of all network resources. For instance, it has to react to the fluctuation of the load and change the assignment of network resources if needed. The key problem is how and when to add and remove resources in order to meet the required Quality of Service (QoS) of users. In this paper, we consider a network based on Software Defined Networking (SDN), Network Function Virtualization (NFV) and Network Slicing technologies. We interest on the management of the deployed network slices. To this end, we propose a Dynamic Handler Framework (DHF) that collects information about slice performances and then determines if the slice resources need to be modified or not. The proposed framework is based on a fuzzy logic algorithm that selects the adequate management decision based on two criteria which are the load ratio and the predicted load-time fairness index. In this paper, a mathematical formulation of our proposal and an evaluation of its efficiency are presented.