Distributed computing architecture, method and device of cloud-side-terminal heterogeneous edge computing network

Abstract

The present invention relates to a distributed computing framework, method and device for a cloud-edge heterogeneous edge computing network. The method includes: determining an energy consumption model; the energy consumption model includes: an edge device model, an edge server model, and a cloud computing center model ; Constructing an objective function based on the energy consumption model; Determining constraints; Normalizing the objective function and the constraints; Grouping the normalized objective functions until each group contains only one objective function and one Corresponding variables; each variable is updated until convergence, and the optimal calculation unloading ratio is obtained. The above-mentioned method in the present invention can obtain an optimal network computing unloading ratio and minimize network energy consumption.

Description

technical field [0001] The present invention relates to the field of heterogeneous edge computing, in particular to a distributed computing architecture, method and device for a cloud-edge heterogeneous edge computing network. Background technique [0002] The heterogeneous edge computing (HetMEC) network in the prior art combines cloud and multi-layer edge computing, so that computing tasks generated at edge devices can be offloaded to servers of different layers for computing, reasonably evacuating computing pressure, and improving computing efficiency , the system delay can be effectively minimized through the joint optimization of computing offloading, computing resource allocation, and transmission resource allocation. However, this solution does not consider the issue of system energy consumption. However, energy consumption is very important in edge computing networks, especially for edge devices. [0003] At present, most of the existing solutions only consider the ...

AI Technical Summary

Problems solved by technology

[0004] Most of the current schemes are centralized optimization algorithms. In the schemes for distributed networks, games are usually used to coordinate the decision-making behavior of independent individuals of various interests. The scalability is insufficient, and it is not suitable for large-scale network large-scale task processing.

Images