网络集群部署数学建模设计与仿真  

Mathematical Modeling Design and Simulation of Network Cluster Deployment

作　　者：徐成桂 徐广顺[1] XU Cheng-gui;XU Guang-shun(Engineering&technical college,Chengdu University of Technology,Basic Teaching Department,Leshan Sichuan 614000,China)

机构地区：[1]成都理工大学工程技术学院,四川乐山614000

出　　处：《计算机仿真》2023年第4期392-396,共5页Computer Simulation

摘　　要：为解决当前网络集群部署系统存在的资源利用率低、部署耗时长等问题,构建基于约束最大熵的网络集群部署数学模型。从并发用户个数、文件储存容量等方面,获取物理主机可提供的资源上限。利用灰色关联分析法评估并发用户个数和数据量间的耦合关系,量化资源需求与集群部署指标。在概率分布集合中选取最大熵模型,运用约束最大熵法估计集群部署指标参数重要性。将网络拟作树结构,计算集群部署代价矩阵,利用图分割方法构建网络集群部署模型。仿真结果表明,所建模型可有效提升网络服务器应用效率,降低网络运行与管理成本。In current network cluster deployment system,resource utilization and deployment time are not ideal.In this paper,a mathematical model of network cluster deployment based on constrained maximum entropy was constructed.First of all,we calculated the upper limit of the resource provided by a physical host from the number of concurrent users and the capacity of file storage.Then we used grey relational analysis to assess the coupling relationship between the number of concurrent users and the amount of data,and thus to quantify resource requirements and cluster deployment indexes.Moreover,we chose a maximum entropy model from the probability distribution set and used the constrained maximum entropy method to estimate the importance of cluster deployment parameters.Finally,we used the network as a tree structure,and calculated the cost matrix of cluster deployment.Finally,we built a model of network cluster deployment by the graph-cut method.Simulation results show that the designed model can effectively improve the application efficiency of network servers and reduce the cost of network operation and management.

关 键 词：约束最大熵 集群部署 模型构建 部署指标参数 图割论 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]