计算机科学 ›› 2011, Vol. 38 ›› Issue (7): 88-92.

• 计算机网络与信息安全 • 上一篇    下一篇

事件驱动的分布式自适应网络管理系统

丁振国,王静懿,赵永杰   

  1. (西安电子科技大学计算机学院 西安710071)
  • 出版日期:2018-11-16 发布日期:2018-11-16
  • 基金资助:
    本文受中央高校基木科研业务费专项资金资助项目((JY10000917003),综合网络管理系统资助。

Event-driven Based Distributive and Adaptive Network Management System

DING Zhen-guo,WANG Jing-y,ZHAO Yong-jie   

  • Online:2018-11-16 Published:2018-11-16

摘要: 借鉴COR13A事件服务的方式,提出并实现了基于事件驱动模型的分布式自适应网络管理系统(EDDM)。该系统在功能上具有强可伸缩性,可实现海量网络事件的缓冲、过滤和分发;同时分布式架构具有良好的自适应性,根据BP神经网络算法可动态调整管理节点的负载,允许管理站的加入和退出,并及时根据实际需要调整系统的管理层次结构。实验结果表明,EDDM可有效缩短故障响应时间,减小网管数据占有的网络带宽。

关键词: 网络管理,分布式,事件驱动,神经网络,自适应

Abstract: Referring to CORI3A event services scheme, the paper proposed an event driven based distributive and adaptive network management system named EDDM The system has very good flexibility in functions, realizing the buffering, filtering, and distributing capabilities of massive network events. Meanwhile distributed architecture has fine adaptability. The load and number of the management nodes can be dynamically ajusted by BP neural network algorithm. Then the management architecture can be changed according to the actual needs. Experiments show that EDDM can effectively reduce the latency of response and the bandwidth occupied by management data.

Key words: Network management, Distributed system, Event-driven, BP neural network, Adaptive system

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!