The development of vehicular services in Internet of vehicles poses challenges for vehicles with limited computation resources to guarantee the quality of service (QoS) of latencysensitive and massive computation onboard services. Vehicular mobile edge computing (VEC) has emerged as an effective technology to enhance vehicular service quality through offloading onboard computation tasks to mobile edge computing (MEC) servers. MEC technology can reduce task processing latency and data transmission latency through its on-premises feature. However, the deployment of VEC still faces several problems such as lacking rational and effective resource allocation schemes. In order to solve these problems, we provide an optimal resource allocation mechanism in vehicular MEC systems (RAVEC) to minimize the total task processing delay among a set of vehicles in a time slot by using a global optimization perspective. The method considers the computation ability of each MEC server at road side unit (RSU) in a road segment, the mobility of each vehicle and the total offloading latency of a set of vehicles to get a best resource allocation plan and achieve onboard task offloading. Simulation results show that RAVEC demonstrates a reliable solution and has a certain value for future research.