Detection of AQM on Paths using Machine Learning Methods

Active Queue Management (AQM) schemes, such as PIE [1] and RED [2], have been particularly effective in combating bufferbloat and attenuating network congestion. Despite their widespread success and improved performance over traditional Tail Drop queuing schemes, the extent of AQM deployment in contemporary networks is not known [3]. Detecting the presence of AQMs in a network has potential to not only provide insight into a network’s internal characteristics (i.e., network tomography), but also facilitate network optimization . In this paper, we address the problem of determining whether a bottleneck router on a given network path is using an AQM or a drop-tail scheme. We assume that we are given a source-to-sink path of interest -along which a bottleneck router existsand data regarding the Round-Trip Times (RTT) and Congestion Window (CWND) sizes with respect to this flow. We develop a reliable classification algorithm that solely uses RTT and CWND information pertaining to a single flow to classify the queuing scheme, Tail Drop or AQM, used by the bottleneck router. We evaluate our method and present results that demonstrate our algorithm’s highly accurate classification ability across a wide array of complex network topologies and configurations.