基于多波动尺度的两层包级自适应抽样

在高速Internet链路的流量测量中,包抽样技术能有效减少被测数据量以节省各种可用资源。文章关注如何提高流量突发和波动周期的抽样精度,提出了一种基于多波动尺度的两层包级自适应抽样方法。该方法能够根据当前流量的突发频度和波动性动态调节每包抽样概率,调节粒度精细。通过在真实流量跟踪下与静态随机抽样方法的对比,两层自适应抽样方法显示了较强的实时捕获流量速率的波动变化和改善抽样精度的能力。

Two Phase Packet-Level Adaptive Sampling on Multi-Fluctuation Scale Traffic

In high-speed Internet links,efficient and effective packet sampling techniques for traffic measurement are not only desirable,but increasingly becoming a necessity.In this paper,we propose a two phase packet-level adaptive sampling method on different traffic fluctuation and burst scales.In particular,we address the issue of how to improve sampling accuracy during burst or fluctuation periods.By contrast to static random sampling method in experiments using real traffic trace,we show that the proposed two phase adaptive sampling has great capability to trace traffic fluctuations and improve sampling accuracy through capturing more fluctuation or burst details.