Novel approach for virtual network function requirement prediction

Abstract

Network Function Virtualization (NFV) is gaining popularity among network operators to provide cost effective and dynamic network services. NFV enables faster service by deploying virtual instances of network functions. While serving dynamic and varying traffic demands, network operators can get benefit from knowing the requirement for the number of Virtual Network Functions (VNFs) ahead of time. VNF requirement prediction method mostly depends on the fluctuation of network traffic load. Predicting the required number of VNFs helps the operator to manage network resources in better ways. VNF prediction method is being considered as an interesting research field for researchers. In our study, we propose VNF requirement prediction and forecasting methods based on deep learning algorithms. Different types of network functions like NAT, Traffic Shaper, IPSec VPN, etc. while performed as a VNF, they require different amount of virtual-CPU (vCPU) depending on the traffic throughput. Here, we also propose deep learning based forecast models for vCPU for heterogeneous VNFs. In this thesis, we provide experimental results which show promising accuracy and low error of our models. Network resource management can be benefited enormously from our findings.