Characterizing the Impact of Active Queue Management on Speed Test Measurements

TL;DR

The paper addresses the gap that traditional speed tests, which emphasize peak throughput, fail to reflect user-perceived responsiveness under load. Through a controlled lab study across multiple AQM schemes (CoDel, FQ-CoDel, SFQ) and with and without burst shaping and competing traffic, it shows that speed-test measurements exhibit significant variability in both throughput and latency distributions that depend on AQM policy. End-to-end results and instantaneous throughput analyses reveal that aggregated metrics can mask important dynamics, especially under load ($LUL$) and cross-traffic conditions. The findings highlight the need to calibrate and enrich speed-test platforms with AQM-aware metrics to produce results that better reflect real user experience and to inform policy and regulatory outcomes.

Abstract

Present day speed test tools measure peak throughput, but often fail to capture the user-perceived responsiveness of a network connection under load. Recently, platforms such as NDT, Ookla Speedtest and Cloudflare Speed Test have introduced metrics such as ``latency under load'' or ``working latency'' to fill this gap. Yet, the sensitivity of these metrics to basic network configurations such as Active Queue Management (AQM) remains poorly understood. In this work, we conduct an empirical study of the impact of AQM on speed test measurements in a laboratory setting. Using controlled experiments, we compare the distribution of throughput and latency under different load measurements across different AQM schemes, including CoDel, FQ-CoDel and Stochastic Fair Queuing (SFQ). On comparing with a standard drop-tail baseline, we find that measurements have high variance across AQM schemes and load conditions. These results highlight the critical role of AQM in shaping how emerging latency metrics should be interpreted, and underscore the need for careful calibration of speed test platforms before their results are used to guide policy or regulatory outcomes.

Figures (9)

• Figure 1: Throughput measurements without burst shaping and without cross traffic has lower variability across different AQM algorithms but often underutilizes the link capacity.
• Figure 2: Latency measurements without burst shaping and without cross traffic does not introduce significant latency spikes due to low of queuing delays.
• Figure 3: Throughput measurement with burst shaping without cross traffic achieve higher link utilization and better use of available bandwidth across all AQM algorithms.
• Figure 4: Latency measurement with burst shaping without cross traffic is still stable due to lower competing load on the link.
• Figure 5: Throughput measurement with burst shaping and TCP cross traffic shows higher variability across different AQM algorithms as these become more prominent under load from competing traffic.