一种基于分类改进的LARS调度算法及其动态参数性能分析  

A LARS scheduling algorithm based on classification improvement and its dynamic parameter performance analysis

作　　者：何志强 林永君[1] HE Zhiqiang LIN Yongjun(School of Control and Computer Engineering, North China Electric Power University, Baoding 071009, China Information Management and Engineering Department, Hebei Finance University, Baoding 071051, China)

机构地区：[1]华北电力大学控制与计算机工程学院,河北保定071009 [2]河北金融学院信息管理与工程系,河北保定071051

出　　处：《河北大学学报（自然科学版）》2017年第5期537-544,共8页Journal of Hebei University(Natural Science Edition)

基　　金：河北省高校智慧金融应用技术研发中心资助项目(XZJ2017006)

摘　　要：以工业以太网与商用网络信息整合为研究背景,从调度算法入手,研究了基于LARS调度算法在多业务网络中改进网络服务质量的问题.在研究LARS调度算法的原理和优缺点基础上,提出了利用数据分类技术改进LARS的方法,并在算法参数不同取值的条件下对转发性能做了测试,证明该方法在实际应用中能够保留LARS的基本特性,同时能够有效提升LARS的执行效率,并具有较好的动态参数稳定性.Aiming at the demand of industrial ethernet and office network information integration,this paper studies the problem of improving the quality of network service in multi-service network based on LARS scheduling algorithm.The principles,advantages and disadvantages of LARS are studied,and a method for improving LARS by using data classification technology is proposed.The forwarding performance test under different parameters proves that the method can preserve LARS＇s basic characteristics,can effectively improve the efficiency of the implementation of LARS,and has a good robustness.

关 键 词：调度算法 分类 排队 时延 

分 类 号：TP393.0[自动化与计算机技术—计算机应用技术]