摘要
针对高速链路中流量测量缺乏可扩展性的问题,提出一种在线挖掘频繁流的算法。通过采用"滑动窗口"机制,构造流抽样函数,自适应地设置抽样门限的方法,实现流大小的无偏估计。基于实际的互联网数据进行仿真实验,结果表明,该算法在保证准确性的同时,具有自适应性和资源可控性。
Aiming at the deficiency of traffic measurement lacking scalability in high-speed link, this paper develops an algorithm to mine frequent flows over online packet stream. Adopting the mechanism of sliding window, the algorithm constructs a sampling function and configures adaptive thresholds to achieve unbiased estimates of flow size. Experiments are conducted based on real network traces. Results demonstrate that the proposed method can achieve adaptability and controllability of resource consumption without sacrificing accuracy.
出处
《计算机工程》
CAS
CSCD
北大核心
2010年第3期122-124,共3页
Computer Engineering
基金
国家"863"计划基金资助项目"目标导向课题"(2007AA01z2a1)
关键词
流量测量
滑动窗口
自适应门限
资源可控性
traffic measurement
sliding window
adaptive threshold
resource constraints
