Cloud-Native Based Task Scheduling and Resource Allocation for Internet of Vehicles

Mobile Edge Computing (MEC) technology has been introduced to the Internet of Vehicles (IoV) to provide the sufficient computing resource. However, designing an efficient task offloading and resource allocation strategy in a distributed MEC system is challenging. To address this issue, this paper proposes a cloud-native based task scheduling and resource allocation strategy. First, we introduce the cloud-native based Vehicular Edge Computing Framework (CVEF), which ensures fast task response and efficient system execution through the collaboration of multiple edge clusters. Next, we jointly formulate task scheduling and resource allocation to minimize total task execution latency and energy consumption, taking into account network features, task properties, microservice characteristics, and computation resource constraints of the cluster. Finally, we model the problem as a Markov game and design a MADDPG based adaptive task scheduling and resource allocation algorithm, where the edge clusters make decisions in a decentralized manner. Numerical simulation results demonstrate the effectiveness of our proposed solution in reducing time and energy consumption.