Dynamic Node Link Model of Hierarchical Edge Computing

Abstract

With the rise of the Internet of Things, edge computing has become one of the key technologies in Internet of Things solutions. In the context of the Industrial Internet of Things, hierarchical edge computing shows its advantages. This article focuses on hierarchical edge computing in the industrial Internet of Things scene, and studies the dynamic resource allocation of hierarchical edge computing networks. When using a hierarchical edge computing network with existing equipment, it is difficult to make changes to existing equipment. Therefore, this article uses queuing theory modeling analysis and proposes Dynamic Link Model based on Nodes Relation. Aiming at the hierarchical edge computing network, this model uses a method based on node connection relationship transfer to achieve load balancing of task flow and completes the dynamic allocation of computing resources in the network, and proposes a time experienced priority queue offloading strategy. The paper uses Java to achieve a dynamic link model experiment based on the connection relationship between nodes. The results show that this scheme has significant advantages in the global average delay of the system, and ensure the loss probability is reasonable within a certain limit.