Devising a method for managing computing resources in a fog layer of the mobile high-density internet of things

Abstract

This study considers a process that manages the distribution of computing resources in the fog layer of the mobile high-density Internet of Things. The task addressed is to reduce the load imbalance of fog servers by devising a method for controlling computing resources in the fog layer when processing information flows. Information flows are formed by intelligent gateways of the mobile high-density Internet of Things, which receive data from the boundary layer. In the process of research, a mathematical model for the process of controlling computing resources in the fog layer was built. Its main difference from existing ones is a module hierarchical structure according to the basic levels of decision-making when managing computing resources. When constructing the model, the principle of process decomposition into adjacent time intervals was used. Its application made it possible to carry out local optimization of the process of managing computing resources in short time intervals. The mathematical model has made it possible to devise a method for controlling computing resources in the fog layer. The main difference of this method from existing ones is that the process optimization is carried out according to the area of the relative deviation from the balanced load in the time interval under study. In addition, a two-stage algorithm for distributing tasks of free fog devices across fog layer servers is also used. That made it possible to reduce the time for finding an approximate solution for distributing computing resources of fog servers by up to 50%. The research results can be attributed to the combined use of the simulated annealing algorithm and the genetic algorithm. The method is effective when the load on the fog layer is from 20% to 70% of the maximum possible load.