With the rapid development of Internet of Things technology and interactive applications, the number of terminal devices in the network is increasing, and the development of interactive applications is hindered by network delay. To solve the network delay, bandwidth, and workload requirements in the new era, edge computing came into being. Edge computing aims to implement computing, storage, communication, and other services at the edge of the network by sinking cloud services from the core network to the edge of the network.Current research studies pay less attention to the impact of edge server location on the system performance, and edge server deployment is one of the key technologies for mobile edge computing. Therefore, we take 5G macrocellular/microcellular cluster as the edge server deployment scenario, propose an equivalent bandwidth-based deployment strategy, establish a mathematical model for edge server deployment, and contract a task experience function as an evaluation index from two aspects: task time and energy overhead. Based on the analysis of the experimental results, it is verified that the deployment strategy based on equivalent bandwidth is superior to other deployment strategies in terms of terminal device task overhead.

Vehicular edge computing (VEC) offers a new paradigm for improving vehicular services and enhancing vehicle capabilities. In the VEC environments, vehicles can offload tasks to edge servers deployed in roadside units or other vehicles via wireless networks to enhance their computation capability. However, the high mobility of vehicles will lead to constant change of the network topology among resources, which will affect the completion time and the reliability of offloaded tasks. Therefore, we proposes a reliability‐based offloading strategy by considering task attributes, resource state, completion time, and reliability to map task flows to resources that can satisfy the time constraints with high reliability. The experimental results show that, compared with existing typical algorithms, the proposed strategy not only reduces the average completion time of tasks, but also ensures the reliability of tasks.

No abstract