| 摘要: |
| 随着网络技术的快速发展以及各种新网络业务的出现，网络的规模日益扩大和复杂。同时，各种网络服务对于网络可靠性的要求也逐渐提高，这就对网络的维护提出了更高的要求。在这种情况下，传统的网络维护模式逐渐不能适应现代网络发展的需求。针对现代大型网络的故障诊断问题，利用大数据分析技术，设计并完成了一套整体的智能化故障诊断系统。在此系统中，针对不同的故障类型，利用能够获取到的数据类型，针对性地设计了相关的故障诊断模块和算法。系统整体基于开源的数据处理平台构建，具备良好的可扩展性。通过实际案例数据的验证，结果表明此系统具备良好的处理实际网络故障的能力，将大数据分析技术应用于网络故障诊断领域也具备较强的可行性。 |
| 关键词: 网络维护 故障诊断 大数据分析技术 开源数据处理平台 |
| DOI： |
|
| 基金项目: |
|
| An intelligent network fault diagnosis system via big data analysis technique |
| ZHU Yongqing,HUANG Xinyu,TANG Hong,CHEN Jiangshan,CHENG Kang,BAO Dewei |
| (Guangzhou Research Institute of China Telecom Co.,Ltd.,Guangzhou 510630,China;Huawei Technologies Co.,Ltd.Nanjing R&D Center,Nanjing 210012,China) |
| Abstract: |
| With the rapidly developed network technology and various new network services,the scale of network becomes larger and more complex.At the same time,various network services require higher network reliability.Under this condition,traditional network maintenance model can′t satisfy the modern network.In this context,this paper designs and develops an intelligent fault diagnosis system for large scale network maintenance.In this system,targeting various types of fault,different diagnosis module and fault diagnosis algorithms are designed and implemented via the corresponding types of data.The system is constructed by open source data processing platform,which has favorable scalability.The validation shows the ability to process the practical network fault and the feasibility of applying the big data analysis technology in network fault diagnosis area. |
| Key words: network maintenance fault diagnosis big data analysis technology open source data processing platform |
