基于IP网络质量优化的监测方案设计

Design of Monitoring Solution for IP Network Quality Optimization

文章从网际互连协议(IP)网络质量出发,针对时延和丢包等网络性能因素对互联网整体质量和行业大客户端到端业务质量的影响,提出了统一的架构设计,结合高速采集(Telemetry)、双向测量(Twamp)、逐流检测(IFIT)和数据流分析(Flow)等新技术的应用,完成了IP网络质量监测及优化方案研究。结果表明,通过文章方案部署可有效提升监测数据的时效性、真实性和可靠性。方案部署后采集海量包粒度的质量数据并分析,借助人工智能(AI)预测能力完成故障预警,在提升了业务质量的同时进一步增强了现有网络“自动驾驶”能力。

In this paper,we consider the Internet Protocol(IP)network quality and study the impact of network performance factors,such as delay and packet loss on the end-to-end service quality of major accounts and overall quality of internet.We propose a unified architecture design and conduct an investigation on IP network quality monitoring and optimization solutions through a combination of Telemetry,Twamp,IFIT,Flow and other applications of new technologies.The results indicate that the deployment of proposed design can provide effective improvements in the timeliness,authenticity and reliability of monitoring data.Meanwhile,it will collect and analyze the quality data of massive packet granularity,and realize fault warning with the help of AI intelligent prediction capabilities after the deployment of design.The proposed design can both improve the quality of services,and further enhance the existing network's"autonomous driving"capabilities.