A Dual−Population Strategy Based Multi−Objective Yin−Yang−Pair Optimization for Cloud Computing

In order to improve the performance of cloud computing, the multi−objective optimization problems in this field need to be solved efficiently. This paper proposes a novel Dual−Population strategy based Multi−Objective Yin−Yang−Pair Optimization which is termed as DP−MOYYPO. The proposed DP−MOYYPO algorithm makes the following three improvements to Front−based Yin−Yang−Pair Optimization (F−YYPO). First, a population of the same size to explore non−cornered points with large crowding distances is added, so as to enhance the uniformity of the optimized individuals’ distribution. Second, the updating method of point P in the splitting stage is modified to reduce the number of function evaluations for the purpose of improving the convergence speed and accuracy. Third, the updating method of P2i in the archive stage is improved. The proposed DP−MOYYPO algorithm has been evaluated using two test suites of the PlatEMO platform. DP−MOYYPO is first compared with F−YYPO and all improved algorithms, and then with five other representative multi−objective optimization algorithms. Experimental results show that, DP−MOYYPO can obtain Pareto fronts with better distribution than F-YYPO, while having faster convergence and higher computational accuracy. And compared with other representative multi-objective optimization algorithms, DP-MOYYPO ranks first in terms of the number of winning test instances in both the convergence metric GD and the integrated performance metric IGD, showing a strong competitive advantage.