Approximating Communication Cost for NFV-Enabled Multicasting

Abstract

Network function virtualization (NFV) and service function chains (SFCs) effectively improve the flexibility of network service provisioning and increase the extent to which scaling can be done. However, finding an efficient deployment of virtual network functions (VNFs) for steering service function chain (SFC) requests is an NP-hard problem. The objective of our study is to obtain an optimal communication cost in VNF deployments and to allow for effective traffic steering in NFV enabled multicasting, when the number of SFC requests is large. Specifically, we have first formulated the problem and proved that it is NP-hard. We then present a 6$\alpha$ approximation algorithm for the centralized approach, where $\alpha$ is the approximation factor, and a O(logn) approximation algorithm for the distributed approach, where n represents the number of nodes. Through extensive simulations on synthetic and real-world networks, we have evaluated the performance of our proposed approach on multi-cast traffic to be better by 18.06% as compared to current state-of-the-art algorithms.