流量测量的关键技术分析与研究*

流量测量在流量计费、网络资源优化和异常检测等方面有广泛的应用。针对传统的流量测量模型缺乏可扩展性的缺点，提出了一种适用于高速网络的可扩展的流量测量模型。该模型采用报文分批处理的思想引入了两级缓冲区结构，使得缓存报文和流量统计两个过程同时进行。基于此模型，抽象了流量测量的两大关键技术，即流量抽样测量技术和概要数据结构。运用此模型进行高速网络的流量测量，不仅会降低资源的需求、提高分析的速度，而且还不会失去准确性。

Analysis and research on key techniques of traffic measurement

Traffic measurement is an important component of network applications including usage-based charging, network optimization and anomaly detection. Based on the research of three typical flow mea-surement techniques, this paper presented a novel scalable flow measurement model deployed in high-speed network to circumvent the deficiency of traditional flow measurement model. Introducing the idea of processing in batches, the model designd two-stage buffer structure to conduct buffering packets and measuring traffic simultaneously. Referred flow sampling technique and referred synopsis data structure as two key techniques in the light of the above model. Implementing the model to measure traffic can not only meet the demand of resources, but also improve the analytical speed without sacrificing accuracy.