A hybrid heuristic computation offloading algorithm based on energy fairness
Computation offloading schemes in edge computing scenarios often face problems,such as uneven resource allocation,unreasonable task assignment and inefficiency.Therefore,we propose a hybrid heuristic computation offloading algorithm based on energy fairness.Specifically,considering the energy fairness,communication resources and computing resources of edge servers,we formulate an optimization problem that minimizes the total energy consumption for completing all tasks.Firstly,we incorporate the energy fairness metric into the selection criteria for the target edge server to obtain the optimal target server.Secondly,we design a hybrid heuristic computation offloading decision algorithm that combines the genetic algorithm(GA)and the simulated annealing(SA)algorithm,thus we could solve the formulated mixed-integer nonlinear programming problem.This algorithm improves the speed of convergence and search quality,avoids getting trapped in local optimum,reduces dependency on initial parameters and enhances algorithm stability.Finally,simulation results validate that the proposed mechanism has significant advantage in terms of energy consumption compared to other benchmark methods and achieves the highest energy fairness for edge servers.It also demonstrates the convergence speed advantage of the proposed mechanism.
万方数据
维普
