基于模糊聚类和改进GRNN的云服务器聚类算法的研究与实现

IVCE cloud platform has thousands of cloud servers and a variety of tasks, a task every half an hour to the cloud server platform will perform, in this case, the cloud server cluster hierarchy is the basis of reasonable task scheduling, the maximum efficiency of the cloud platform. Because of its strong nonlinear mapping ability and flexible network structure GRNN algorithm suitable for solving nonlinear problems in cluster server, but belongs to the GRNN network prediction algorithm, is proposed in this paper based on the fuzzy C means algorithm Research and Implementation of Cloud Server Clustering Algorithm Based on Fuzzy Clustering and Improved GRNN form the server clustering algorithm, and add improved fruit fly optimization algorithm can overcome GRNN slow convergence and local minimum problem, and IVCE platform to collect 23 performance indexes, this paper will FCM the output of the algorithm through the neighborhood rough set analysis algorithm for feature selection, avoid the dimension disaster. The research results of this paper compared to the existing algorithm, the speed is accelerated by 1.43 and 3.22 respectively and 3.72 times; the step number, were also reduced by 1.61, 5 and 6 times, and in the operation of the algorithm, the machine\'s memory and CPU usage control in about 50%, does not affect the normal function. Moreover, after the algorithm clustering in this paper, the task scheduling of the platform is more balanced.