Research On Load Balancing Technology In Edge Computing

With the development of emerging technologies such as the Internet of Things(IoT),a large number of devices are connected to the network,and the safe and effective processing of massive data has become an urgent problem to be solved.Cloud computing is often deployed in the center network which is far from the end-users,and it is difficult to meet the demand for mobility support,geo-distribution,location awareness,and low latency.To overcome these shortcomings,a new edge computing model is proposed.Edge computing extends cloud computing services to the edge of the network,which can further shorten the information transmission and reduce processing delay.However,edge computing nodes are usually heterogeneous and have limited computing capability.Making full use of edge computing resources,reasonably realizes the data storage,task allocation,and resource scheduling,i.e.,realizes the load balancing in edge computing has become one of the hot issues among researchers.In this paper,based on the edge computing framework,we investigate the problem of load balancing in edge computing.The specific research contents are as follows.(1)The energy-saving load balancing task allocation algorithm in edge computing is designed.For the problem of perfect global information in edge computing,we propose an edge computing network architecture model based on intermediary nodes.To deal with the task completion time and the energy consumption of the edge computing layer during task processing,an energy-saving load balancing task allocation algorithm is designed.The algorithm introduces slack variables to transform the constraint problem into an unconstrained problem and uses the generalized Lagrange multiplier method to solve the problem.Finally,numerical simulations are executed to validate the effectiveness and correctness of the proposed algorithms.It is shown that the proposed algorithms can make a trade-off between the user's demands for task completion time and system energy consumption by adjusting weight properly.Apart from improving the user experience,it can effectively reduce the energy consumption of edge nodes and extend the network life of the edge computing system.(2)The load balancing task scheduling algorithm based on an improved genetic algorithm is designed.This paper proposes an edge computing load balancing algorithm(Improve Adaptive Genetic Algorithm Load Balancing,IAGALB)based on an improved adaptive genetic algorithm for many indivisible tasks.Different from the traditional roulette selection method,the algorithm of this paper selects the chromosomes with lower fitness values to enter the next generation by sorting the fitness function values.A new crossover probability and mutation probability are proposed in the crossover and mutation operations,which can effectively maintain the stability of the population,retain good individuals and genes.The experimental results show that the proposed algorithm has better convergence and stability,reduces the task completion time effectively,and obtains a better task scheduling scheme.