光纤网络信道边界节点均衡负载控制研究
|
摘要
在光纤网络通信信道发生阻塞的情况下,易造成边界节点负载失衡,导致通信执行时间长且丢包严重。因此,提出光纤网络信道边界节点均衡负载控制方法。采用任务调度服务器分配节点任务,将运算结果反馈给相关的客户端,对获取的阻塞节点进行分配时,需要考虑节点的优先级量化方式,结合光纤网络节点的可用内存、CPU主频、正在运行的节点数以及等待运行的节点数四个相关因素,获得节点优先级,优化节点分配精度,最后定义节点队列,构建光纤网络信道边界节点负载均衡处理模型,完成对光纤网络信道边界节点负载均衡化处理。实验结果证明,该方法对阻塞节点负载均衡化处理过程中节点原数据得以更大程度保留,其任务响应和执行时间较短且丢包率低,光纤网络阻塞节点负载均衡化处理效果明显。
When the communication channel of optical fiber network is blocked,it is easy to cause the load imbalance of the boundary node,which leads to the long time cost of communication execution and the serious packet loss.Therefore,a load balancing control method for fiber channel boundary nodes is proposed. The task scheduling server is used to allocate node tasks,and the results are fed back to the relevant clients. When allocating the acquired blocking nodes,we need to consider the priority quantization mode of the nodes and combine the available memory of the fiber optic network nodes. The main frequency of CPU,the number of running nodes and the number of nodes waiting to run are four related factors. The priority of nodes is obtained,the accuracy of node allocation is optimized,and the queue of nodes is defined. The load balancing model of the fiber channel boundary node is constructed,and the load balance of the fiber channel boundary node is completed. The experimental results show that the proposed method can preserve the original data of the blocking nodes to a greater extent in the process of load balancing,and the task response and execution time is shorter,the packet loss rate is low,and the load balance processing effect of the blocking nodes in optical fiber networks is obvious.
When the communication channel of optical fiber network is blocked,it is easy to cause the load imbalance of the boundary node,which leads to the long time cost of communication execution and the serious packet loss.Therefore,a load balancing control method for fiber channel boundary nodes is proposed. The task scheduling server is used to allocate node tasks,and the results are fed back to the relevant clients. When allocating the acquired blocking nodes,we need to consider the priority quantization mode of the nodes and combine the available memory of the fiber optic network nodes. The main frequency of CPU,the number of running nodes and the number of nodes waiting to run are four related factors. The priority of nodes is obtained,the accuracy of node allocation is optimized,and the queue of nodes is defined. The load balancing model of the fiber channel boundary node is constructed,and the load balance of the fiber channel boundary node is completed. The experimental results show that the proposed method can preserve the original data of the blocking nodes to a greater extent in the process of load balancing,and the task response and execution time is shorter,the packet loss rate is low,and the load balance processing effect of the blocking nodes in optical fiber networks is obvious.
引文
[1] CHEN H,GAN C,YIN M,et al.Novel Scheme of Carrier Tri-reuse and Distribution Fiber Protection Based on Round Shift Method in Optical Access Network[J].Journal of Optical Communications,2015,36(2):1-5.
[2] BAYRAM I S,MICHAILIDIS G,DEVETSIKIOTIS M.Unsplittable Load Balancing in a Network of Charging Stations Under Qo S Guarantees[J]. IEEE Transactions on Smart Grid,2015,6(3):1292-1302.
[3]吕淑芳.无线传感器网络节点定位研究综述[J].传感器与微系统,2016,35(5):1-3.
[4]路士增,姜刚顺,隋青美,等.基于小波变换和支持向量多分类机的光纤布拉格光栅低速冲击定位系统[J].中国激光,2014(3):131-137.
[5]郑秀.基于半导体光放大器的波长转换及集成芯片的基础研究[D].成都:电子科技大学,2017,38(4):132-136.
[6]杨正华.基于微纳光纤的黑磷全光调制和波长转换研究[D].深圳:深圳大学,2017,52(10):2239-2246.
[7]巩稼民,刘建花,沈一楠,等.基于掺锗光子晶体光纤的可调谐多波长转换器[J].半导体光电,2018,39(1):15-20.
[8]王丽娟,李锦明,杜东海.基于FPGA的SFP光纤通信控制器的研究与设计[J].电子器件,2016,39(3):605-610.
[9] BI F,ZHENG T,QU H,et al.A Harmful-Intrusion Detection Method Based on Background Reconstruction and TwoDimensional K-S Test in an Optical Fiber Pre-Warning System[J].Photon sensor,2016,6(2):143-152.
[10]石应波,王勇,邰仁忠,等.高分辨宽能段软X射线光束线热负载影响与修正[J].激光与光电子学进展,2014,51(7):194-200.
[11]董毅,赵尚弘,李勇军,等.基于蚁群算法的分布式卫星光网络波长路由分配技术研究[J].电子与信息学报,2015(11):2650-2656.
[12]赵清华,张振振,赵宁波,等.优化光纤铒离子分布实现少模掺铒光纤放大器模式增益均衡[J].激光与光电子学进展,2016(3):55-61.
[13]张惠忠,董泽.相干光通信中预均衡技术的激光器线宽容忍度[J].激光与光电子学进展,2017(3):102-107.
[14]佚名.光纤网络中的哈密尔顿路径[J].中国光学,2014(02):340-340.
[15]刘宏月,梁大开,韩晓林.基于长周期光纤光栅横向负载特性的混凝土结构钢筋锈蚀监测[J].光学学报,2013,33(4):22-28.
[16]钟平,谭敏,刘富国,等.一种应用于非亮度均衡干涉条纹的提取方法[J].应用激光,2015(6):718-723.
[17]曾祥烨,刘剑飞,赵启大.偏振复用光纤通信系统色散均衡器及算法的研究[J].光学学报,2012,32(2):69-73.
[18]陈丹,柯熙政,乔薇.基于子空间的无线光通信副载波盲均衡算法研究[J].红外与激光工程,2015,44(8):2528-2534.
[19]张龙,倪国强,裴京,等.信号波形调制多阶光盘信道部分响应多项式与均衡器的设计[J].光子学报,2013,42(11):1360-1364.
[20]王忠东,张小文,赵承强,等.变电站光纤通信中的最优节点选取模型设计[J].科学技术与工程,2018(2):264-269.
[2] BAYRAM I S,MICHAILIDIS G,DEVETSIKIOTIS M.Unsplittable Load Balancing in a Network of Charging Stations Under Qo S Guarantees[J]. IEEE Transactions on Smart Grid,2015,6(3):1292-1302.
[3]吕淑芳.无线传感器网络节点定位研究综述[J].传感器与微系统,2016,35(5):1-3.
[4]路士增,姜刚顺,隋青美,等.基于小波变换和支持向量多分类机的光纤布拉格光栅低速冲击定位系统[J].中国激光,2014(3):131-137.
[5]郑秀.基于半导体光放大器的波长转换及集成芯片的基础研究[D].成都:电子科技大学,2017,38(4):132-136.
[6]杨正华.基于微纳光纤的黑磷全光调制和波长转换研究[D].深圳:深圳大学,2017,52(10):2239-2246.
[7]巩稼民,刘建花,沈一楠,等.基于掺锗光子晶体光纤的可调谐多波长转换器[J].半导体光电,2018,39(1):15-20.
[8]王丽娟,李锦明,杜东海.基于FPGA的SFP光纤通信控制器的研究与设计[J].电子器件,2016,39(3):605-610.
[9] BI F,ZHENG T,QU H,et al.A Harmful-Intrusion Detection Method Based on Background Reconstruction and TwoDimensional K-S Test in an Optical Fiber Pre-Warning System[J].Photon sensor,2016,6(2):143-152.
[10]石应波,王勇,邰仁忠,等.高分辨宽能段软X射线光束线热负载影响与修正[J].激光与光电子学进展,2014,51(7):194-200.
[11]董毅,赵尚弘,李勇军,等.基于蚁群算法的分布式卫星光网络波长路由分配技术研究[J].电子与信息学报,2015(11):2650-2656.
[12]赵清华,张振振,赵宁波,等.优化光纤铒离子分布实现少模掺铒光纤放大器模式增益均衡[J].激光与光电子学进展,2016(3):55-61.
[13]张惠忠,董泽.相干光通信中预均衡技术的激光器线宽容忍度[J].激光与光电子学进展,2017(3):102-107.
[14]佚名.光纤网络中的哈密尔顿路径[J].中国光学,2014(02):340-340.
[15]刘宏月,梁大开,韩晓林.基于长周期光纤光栅横向负载特性的混凝土结构钢筋锈蚀监测[J].光学学报,2013,33(4):22-28.
[16]钟平,谭敏,刘富国,等.一种应用于非亮度均衡干涉条纹的提取方法[J].应用激光,2015(6):718-723.
[17]曾祥烨,刘剑飞,赵启大.偏振复用光纤通信系统色散均衡器及算法的研究[J].光学学报,2012,32(2):69-73.
[18]陈丹,柯熙政,乔薇.基于子空间的无线光通信副载波盲均衡算法研究[J].红外与激光工程,2015,44(8):2528-2534.
[19]张龙,倪国强,裴京,等.信号波形调制多阶光盘信道部分响应多项式与均衡器的设计[J].光子学报,2013,42(11):1360-1364.
[20]王忠东,张小文,赵承强,等.变电站光纤通信中的最优节点选取模型设计[J].科学技术与工程,2018(2):264-269.