强负荷网络环境下的大型负载分析平台的设计与优化  

Design and optimization of large load analysis platform in strong load network environment

作　　者：陈青艳[1] 陈开慧[1] 

机构地区：[1]玉林师范学院,广西玉林537000

出　　处：《现代电子技术》2017年第2期158-162,共5页Modern Electronics Technique

基　　金：国家自然科学基金面上项目(20719662);广西壮族自治区技术研究项目:网络入侵容忍理论及应用技术研究(自筹项目)(2013LX112);玉林师范学院校级课题:基于数据挖掘的电商企业客户关系管理应用研究(2014YJYB02)

摘　　要：强负荷网络环境下,传统大型负载分析平台存在资源占用率高、效率低等问题。因此,设计了基于Web服务器集群的大型负载分析平台,确保在强负荷网络环境下,通过Web服务器集群提供高可用的服务,实现负载均衡。分析了负载分析平台运行流程,基于该流程设计了负载分析平台的总体结构,平台通过接收模块接收用户申请,再采用分类模块对用户申请进行分类,通过分发模块向用户申请分配相应的进程。信息收集模块将不同服务器节点的运行状态反馈给方案模块,促使方案模块融合神经网络和遗传算法对不同节点的资源分配状态进行设置。分析模块对总体负载分析平台中的不同服务器节点进行监测,分析不同节点是否存在故障以及节点的负载是否均衡。软件设计中给出了负载分析平台模块功能结构,以及信息采集代码设计。实验结果表明,所设计负载分析平台的CPU平均利用率和任务平均处理时间都较优,并且具有较好的吞吐量。Since the traditional large load analysis platform has the problems of high resource occupancy rate and low efficiency in strong load network environment,a large load analysis platform based on Web server cluster was designed to insure the improvement of the useable service by Web server cluster and balance the load in the strong load network environment. The running process of the load analysis platform is analyzed. On this basis,the overall structure of the load analysis platform was designed. The user application is received by the receiving module of the platform,and classified with the classification module.The platform assigns the corresponding progress to the use application through the distribution module. The information collection module is used to feed the running status of different server nodes back to the plan module to prompt the fusion of neural network and genetic algorithm for setting the resource allocation status of different nodes. The analysis module is used to monitor the different server nodes in general load analysis platform to analyze whether the different nodes exist the failure and the node loads are balanced. The module function structure of the load analysis platform and code design of the information collection are given in software design. The experimental result shows that the CPU average utilization and task average processing time of the designed load analysis platform are optimal,and the platform has enough throughput.

关 键 词：负载分析平台 服务器节点监测 WEB服务器 平台结构设计 

分 类 号：TN711-34[电子电信—电路与系统] TP302.1[自动化与计算机技术—计算机系统结构]