Offloading Edge Vehicular Services in Realistic Urban Environments

Abstract

The imminent deployment of 5G and the rapid development of multi-access edge computing standards are demanding advances in terms of vehicular low latency offloading design and modelling proposals. In this paper we describe the functionalities of a high-level multi-access edge computing orchestrator that arranges location based vehicular edge services by the means of hierarchical dynamic resource management. In this way low latency responses can be guaranteed due to the geo-aware and energy efficient service allocation and dynamic migration. The first steps towards the definition of a vehicle to infrastructure communication specification are also provided. We study the efficiency of our proposal applied to an infotainment case study deployed in the city centre of Alicante, Spain. The simulation results obtained show that latencies perceived by vehicles generally range from optimal to first order sub-optimal scales all over the coverage area and that the presented offloading solution energetically scales with the number of hosts at the edge.