Task Scheduling Based on Dynamic Non-Linear PSO in Cloud Environment
Jian Chang, Zhigang Hu, Yong Tao, Zhou Zhou · 2018
To effectively increase the performance of the cloud computing system, an important challenge is the scheduling of tasks to processors in order to achieve minimal energy consumption. In this paper, a Dynamic Non-linear modified Particle Swarm Optimization (DNPSO) algorithm was proposed to overcome the problem of local optimality and slow convergence of the standard Particle Swarm Optimization (PSO) by constructing inertia weight function. It's a precise and effective optimization algorithm that is easier to implement than the existing evolutionary algorithms. As is well-known that the task assignment problem is NP-complete, the performance of this model was evaluated by simulation results. The results show that the DNPSO algorithm can effectively reduce the total energy consumption compared with other algorithms.