Optimal Service Placement with QoS Monitoring in NFV and Slicing Enabled 5G IoT Networks

Abstract

Network function virtualization (NFV) and network slicing are two promising enabling technologies for 5G networks. Considering the volume of data traffic generated by Internet of things (IoT) applications and their service requirement diversity as well as that network resources are spread across different locations, it is imperative to find solutions for optimal service placement and resource allocation for quality of service (QoS) provisioning. In this paper, we address the challenges of optimal network service placement with active QoS monitoring in NFV and network slicing enabled 5G IoT networks and propose a network architecture with optimal computation and resource placement over core, local, and edge data centers. The solution is implemented through virtualized infrastructure managers where operation costs and QoS requirements are considered for service placement. Optimal algorithms are developed based on a control system hub platform with an open source management and orchestration framework. To monitor the performance during traffic runtime, virtual charmed factors are adopted for control and QoS measurement.