Load balancing method for Internet of Vehicles edge calculation

Abstract

The invention relates to a load balancing method for Internet of Vehicles edge calculation, and belongs to the field of Internet of Vehicles. The method comprises the steps that an MEC server is initialized; the MEC server periodically updates the index weight, calculates the performance and the load rate of the server, judges the load state of the server and uploads related information to a dispatching center; the dispatching center receives the information uploaded by each MEC server, and performs table storage, updating and other operations on the information; if the scheduling center receives task migration requests of the servers, the servers are classified, the servers capable of receiving the migration tasks are screened out, and the transition probability of each server is calculated; the scheduling center arranges task migration according to the transition probability, an eNodeB receives scheduling arrangement of the scheduling center and transmits the task to the target eNodeB through an X2 port, and after task calculation is completed, a calculation result is returned; and a next cycle is waited for. The load balance degree of the server cluster can be effectively improved, and the task completion time is shortened.

Description

technical field [0001] The invention belongs to the field of the Internet of Vehicles, and relates to a load balancing method for edge computing of the Internet of Vehicles. Background technique [0002] With the continuous development of the Internet of Vehicles, C-V2X (Cellular-Vehicle to Everything) technology based on cellular communication systems and MEC (Mobile Edge Computing) technology based on multi-access edge computing are getting closer and closer. C-V2X technology can greatly reduce The cost of deploying autonomous driving and Internet of Vehicles in the future, while MEC technology migrates computing, storage, and business service capabilities to the edge of the network close to the terminal or data source, and has the characteristics of localized processing and distributed deployment. [0003] The integration of C-V2X technology and MEC technology can realize the collaborative interaction of "human-vehicle-road-cloud". However, on the road, due to the uneven...

AI Technical Summary

Problems solved by technology

[0006] The traditional load balancing method focuses on exploring the task allocation strategy, while ignoring issues such as when the server performs task allocation and how to accurately judge the server's operating status. These issues have a great impact on the allocation of subsequent tasks, especially in the Internet of Vehicles environment. Due to the mobility of vehicles, the load status of each MEC server is quite different, which can easily lead to problems such as low cluster load balance and low resource utilization.

Images