End-to-End Network Slice Resource Allocation for 5G Mobile Edge Computing

Mobile Edge Computing (MEC) is a key feature in future 5G mobile networks. It enables, low latency services, by bringing them closer towards the users. In 5G also a new logical network architecture providing isolated logical Network Slices (NSLs) is introduced. This necessitates an initial capacity planning and allocation of the virtual NSL resources on a shared end-to-end mobile network infrastructure. The Network Slice Embedding (NSE) on a complex mixed wired and wireless mobile network infrastructure with MEC has been recognized in literature. However, there is no comprehensive solution for end-to-end NSL embedding taking the possibilities of low latency services by using MEC into account. In this paper, an efficient, exact approach for NSE in 5G mobile networks is proposed. Based on an Integer Linear Program (ILP) formulation of the NSE problem, the optimal set of NSLs, in terms of NSL revenue and cost, is determined. The required NSL applications, functions and services are allocated nearly optimally on the 5G end-to-end mobile network infrastructure. The presented solution also provides the optimal number of application instances and their optimal locations on the Mobile Edge Cloud (ME), even for multiple User Equipment (UE) connectivity. Extensive evaluation shows that the proposed holistic approach for NSE and ME allocation is feasible and efficient for small and medium sized problem instances.