An effective metaheuristic algorithm for the deployment problem of edge computing servers

Since various sensors and appliances will be installed in the near future, these devices usually accompanies a large number of computation needs. The centralized computing is not the only solution in this new network environment. Instead of centralized computing, distributing the computing tasks to edge computing servers to further reduce the latency and costs and mitigate the computing loading of the central server is a potential solution. How to deploy the edge computing servers in such network environment has become a promising research issue in recent years, and therefore a novel metaheuristic algorithm is presented in this paper to deploy the edge computing servers. Different from other metaheuristic algorithms, the proposed method will divide the "search space" into a certain number of subspaces and assess the investment of each subspace based on the potential to determine whether to increase or decrease the computation resource of a subspace during the convergence process. The experimental results show that the proposed algorithm can find a better result than all the other search algorithms compared in this paper for solving the deployment problem of edge computing servers.