A cloud computing oriented neural network for resource demands and management scheduling

被引：3

作者：

Lou, Gaoxiang ^{[1
]}

Cai, Zongyan ^{[1
]}

机构：

[1] School of Construction Machinery, Chang'an University Xi'an, Shaanxi,710064, China

来源：

International Journal of Network Security | 2019年 / 21卷 / 03期

关键词：

Cloud resource scheduling - Memory utilization - Particle swarm algorithm - Radial base function - Radial base function neural networks - RBF Neural Network - Resource management and scheduling - Resource-scheduling;

D O I：

10.6633/IJNS.20190521(3).14

中图分类号：

学科分类号：

摘要：

Cloud computing, a new kind of resource sharing service system, can provide virtual resource services such as infrastructure and platform for users who access it through the Internet. Its service quality is related to resource management and scheduling. In this study, CloudSim3.0 simulation platform was used as a simulation platform for cloud computing resource scheduling to test the performance of radial base function (RBF) neural network based on particle swarm optimization (PSO) and RBF neural network based on Improved Particle Swarm Optimization (IPSO) in cloud resource scheduling and configuration. The results showed that the CPU and memory utilization rate and processing time of the two algorithms increased with the increase of processing tasks. It was found that compared to PSO-RBF, IPSO-RBF had higher CPU and memory utilization rate and shorter processing time and converged faster and found the best position of particles after only 30 iterations with small uctuation amplitude. In addition, IPSO-RBF had better performance in balancing the load of different kinds of physical resources compared to PSO-RBF. © 2019, International Journal of Network Security.

引用

页码：477 / 482