New SIGCOMM Study on content-based differentiation is out

During ACM SIGCOMM 2019 in Beijing on 19-24 August 2019, an interesting analysis on net neutrality, with focus on US-based ISPs was presented. ACM SIGCOMM is the flagship annual conference of the Association of Computing Machinery (ACM) Special Interest Group on Data Communication (SIGCOMM) on the applications, technologies, architectures and protocols for computer communication. The document sums up an extensive one-year study of content based traffic differentiation policies deployed in operational networks, using result from 1,045,413 crowdsourced measurements conducted by 126,249 users across 2,735 ISPs in 183 countries/regions.

Net neutrality has been the subject of considerable public debate over the past decade. Despite the potential impact on content providers and users, there is currently a lack of tools or data for stakeholders to independently audit the net neutrality policies of network providers.

The authors of the study, from the Northeastern University and the University of Massachusetts Amherst, have developed and evaluated a methodology that combines individual per-device measurements to form high-confidence, statistically significant inferences of differentiation practices, including fixed-rate bandwidth limits (i.e., throttling) and delayed throttling practices. Using this approach, they identified differentiation in both mobile and WiFi networks, comprising 30 ISPs in 7 countries. They also investigated the impact of throttling practices on video streaming resolution for several popular video streaming providers.

The data were collected through a special application Wehe used in the eponymous project of the Northwestern University. For this study, Wehe uses traces recorded from YouTube, Netflix, Amazon Prime Video, NBC Sports, Vimeo, Spotify, and Skype. The authors selected the first five apps because video streaming is a common target of traffic differentiation; included Spotify because some mobile plans indicate rate limits on streaming audio, and Skype because a telephony app may compete with mobile providers’ voice services.

The team conducting the study gathered sufficient samples to detect differentiation in 144 ISPs, and detected differentiation in 30 of them. This suggests that the majority of ISPs that they studied do not deploy content-based differentiation. They also found that most throttling targets video streaming.

Photo Copyright: Adobe Stock – terovesalainen

Leave a Reply