Research Article
A Hybrid Task Offloading and Service Cache Scheme for Vehicular Edge Computing
@INPROCEEDINGS{10.1007/978-3-031-28990-3_3, author={Linyu Sun and Yancong Deng and Xingxia Dai and Zhu Xiao}, title={A Hybrid Task Offloading and Service Cache Scheme for Vehicular Edge Computing}, proceedings={Edge Computing and IoT: Systems, Management and Security. Third EAI International Conference, ICECI 2022, Virtual Event, December 13-14, 2022, Proceedings}, proceedings_a={ICECI}, year={2023}, month={3}, keywords={}, doi={10.1007/978-3-031-28990-3_3} }- Linyu Sun
Yancong Deng
Xingxia Dai
Zhu Xiao
Year: 2023
A Hybrid Task Offloading and Service Cache Scheme for Vehicular Edge Computing
ICECI
Springer
DOI: 10.1007/978-3-031-28990-3_3
Abstract
The development of 5G, IoT, and other technologies has promoted the emergence of emerging applications, including augmented reality, autonomous driving, and so on. These applications are usually delay-sensitive and energy intensive, which have strict delay constraints and need to consume a lot of computing resources. In order to ensure the quality of service of these applications, this study proposes a framework that combines task offloading and service caching in the local-edge-cloud collaboration system. In order to obtain a satisfactory offloading decision, this study first proposes a distributed task offloading algorithm based on non-cooperative game theory storage which makes the decision of local processing or offloading to the side server with the goal of minimizing system cost over time and energy consumption. Considering the limited storage resources of the edge, this study uses a 0–1 knapsack algorithm to realize dynamic service caching based on task popularity based on the original offloading decision. Based on the results of service caching, if the task initially decides to offload to the edge server and the edge server does not cache the services required by the task, the task will be unloaded locally or to the cloud.
