Heuristic algorithms based on population updating strategies are commonly used for solving multi-objective virtual machine consolidation problems.However,most of these algorithms do not consider the optimization of the initial population,leading to slow convergence.To address this slow convergence issue,this paper takes the genetic algorithm as an example and proposes a multi-objective sensitive initial population generation method,with the core idea of analyzing the resource characteristics of each target and establishing a constrained model for optimizing the initial population and improving its quality.Experimental results show that this method significantly improves the search performance of the genetic algorithm,and is better than the other two improved strategies.
万方数据