MRoCO: A Novel Approach to Structured Application Scheduling with a Hybrid Vehicular Cloud-Edge Environment

Vehicle Edge Computing(VEC) provides an emerging model for improving vehicle service and provisioning computational resources in motion. Recently, considerable research efforts are paid into the problem of task scheduling and application deployment in a vehicular edge computing environment. Instead of considering monolithic and unstructured tasks, this paper addresses the task deployment problem with multiple computation-intensive vehicle applications unloaded into Roadside Unit (RSU) or cloud in a VEC environment and applications are considered to be structured. Therefore, distributing tasks to different RSU servers for calculation is more reasonable to avoid resource contention. In consideration of the ending time constraints of each application and the inherent dependencies on processing applications, we interpret the scheduling into a combinatorial optimization formulation. We propose an efficient offloading algorithm (MRoCO) through prioritizing a task dependency-aware prioritizing mechanism and applying a hybrid edge cloud-based task mapping mechanism. Numerical results demonstrate that MRoCO can beat other existing approaches in terms of lower averaged application finish time, waiting time and on-time finish rate.