Tournament based equilibrium optimization for minimizing energy consumption on dynamic task scheduling in cloud-edge computing

With the increasing advancements in the Internet of Things (IoT) and the growing production of tasks by IoT devices, the demand for cloud computing centers has become more critical than ever. The energy consumption in cloud computing servers has a significant impact on the overall costs and environmental pollution. This article addresses the task allocation problem to cloud computing servers with the aim of reducing energy consumption in those servers while maintaining Quality of Service (QoS). Evolutionary algorithms have been employed to solve this NP-hard problem. In this paper, a novel version of Equilibrium Optimization algorithm is defined and used for finding good solutions for this problem. In the proposed algorithm, a tournament operator is introduced to control selection pressure and enhance the algorithm’s exploration capability during local optima convergence, added to the EO algorithm. The utilization of this operator in the proposed algorithm eliminates the need for sorting all search agents at each iteration, resulting in reduced execution time. The simulation results indicate that the proposed algorithm has demonstrated a 24