一种基于HNN的云服务组合优化  

A Cloud Service Composition Optimization Based on HNN

作　　者：张会丽[1,2] 李志河 Zhang Huili;Li Zhihe(Linfen Vocational and Technical College,Linfen 041000,China;School of Educational Science,Shanxi Normal University,Linfen 041000,China)

机构地区：[1]临汾职业技术学院,山西临汾041000 [2]山西师范大学教育科学学院,山西临汾041000

出　　处：《系统仿真学报》2019年第11期2335-2343,共9页Journal of System Simulation

基　　金：国家社会科学基金(BIA180202);教育部信息化教学研究课题(2018LXB0179)

摘　　要：随着云服务应用开发的日新月异,如何有效地在云平台上实现优化服务的组合,提升云平台系统的整体性能是一个亟待解决的研究问题。为提升云服务的效率,提出一种基于霍普菲尔德神经网络的组合优化模型。该方法针对云服务问题建模;设计一种带有柯西扰动技术的PSO算法来改进霍普菲尔德模型,将该云服务问题表达为霍普菲尔德神经网络能量模型进行优化。通过实验比较证明,该方法比其它典型算法可以更加有效地提升云服务组合优化执行的效率。With the rapid development of Cloud service application, how to effectively optimize the composition of Cloud services on cloud platform and improve the overall performance of cloud platform system have become an urgent research issue. In order to improve the efficiency of Cloud services, a combined optimization model based on Hopfield neural network is proposed. The problem of Cloud services is modeled. The problem is expressed as Hopfield Neural Network energy model for optimization, and a PSO group algorithm with Cauchy disturbance is designed to improve the Hopfield model. The experimental comparison shows that the method can improve the efficiency of Cloud service composition optimization more effectively than other typical algorithms.

关 键 词：霍普菲尔德神经网络 组合优化 WEB服务 资源约束 负载均衡 云计算 

分 类 号：TP393[自动化与计算机技术—计算机应用技术] TP181[自动化与计算机技术—计算机科学与技术]