Energy-Aware Placement of Network Functions in Edge-Based Infrastructures with Open Source MANO and Kubernetes

The virtualization of network functions aims to replace traditional network functions running on proprietary middleboxes with software instances running on general-purpose virtualization solutions, looking for more flexible, scalable and sustainable networks. However, despite the availability of platforms and technologies that enable its realisation, there are still technical challenges that have to be addressed to obtain those benefits. One of the challenges is the efficient placement of network functions, considering, for instance, the energy footprint among other constraints and available resources. This paper proposes an energy-aware virtual network function placement and resource-allocation solution for heterogeneous edge infrastructures that considers the computation and communication delays according to the virtual network functions’ location in the infrastructure. The solution has been integrated with the ETSI-sponsored project Open Source Management and Orchestration (OSM) as an extension that allows the configuration of virtual network functions and their subsequent resource allocation and deployment at the edge, minimizing energy consumption and ensuring the quality of service. Applied to the deployment of augmented reality services in different scenarios, the results show up to a 51% reduction in energy consumption compared to the default OSM placement and quality of service compliance in all scenarios considered.