论文部分内容阅读
为了进一步降低大流检测算法在高速网络中的漏检率并提高大流流量的测量精度,提出了一种基于LEAST淘汰策略和计数型布鲁姆过滤器(CBF)两级结构的检测算法.在该算法中,CBF只是被用来滤除网络中的小流,并不须要占用太多的缓存空间.而通过CBF的流将进入下一级过滤机构中按LEAST淘汰策略进一步地筛选.从理论上分析了该算法对大流的检测能力,并针对其不足,提出了时间窗口和预留函数两种优化机制.最后基于实际的流量数据进行了实验验证,结果表明该算法的各项评价指标均优于同类算法.
In order to further reduce the missed detection rate of high-flow detection algorithm in high-speed network and improve the measurement accuracy of large-flow, a detection algorithm based on LEAST elimination strategy and two-stage structure of counting type Bloom filter (CBF) is proposed. In this algorithm, CBF is only used to filter out the small stream in the network, does not need to take up too much buffer space, while the stream through the CBF will enter the next level of filtering institutions by LEAST knockout strategy to further screen from In order to overcome the shortcomings of the proposed algorithm, we propose two optimization mechanisms: time window and reservation function, and finally verify the algorithm based on the actual traffic data. The results show that the algorithm’s evaluation Indicators are better than similar algorithms.