Abstract
Vehicular fog computing (VFC) has been proposed as a promising solution to overcome the limitations of edge computing. In VFC, the idle resources of moving and parked vehicles can be used for compute-intensive applications of resource-limited vehicles by offloading their tasks to them. For this to succeed, selecting an appropriate target fog node needs to consider various constraints. This paper argues that the selection process should broadly follow the steps needed to form a service level agreement (SLA) to ensure that the right target fog node is selected. We identify the different requirements that need to be addressed in forming such a SLA before surveying the existing literature to determine if the existing approaches of task offloading in VFC address them or not. Based on the analysis, we conclude the paper by discussing open gaps that need to be addressed for efficient task offloading in VFC.
•The selection of an appropriate target fog node in Vehicular Fog Computing’s (VFC) is a complex process.•For its success, we argue that the selection process should broadly follow the steps needed to form an SLA.•We define three requirements (R1–R3) to consider in selecting the right target fog node by managing the different constraints.•We then study the existing approaches for task offloading in VFC and determine their shortcomings in forming an informed SLA.