Quantifying the Influence of Regulatory Instructions over the Detection of Network Neutrality Violations

Abstract

The state-of-the-art solutions for detection of Network Neutrality (NN) violations assume that all detectable Traffic Differentiations (TDs) are in fact NN violations. However, legislators and regulatory agencies state instructions that establish which TDs may be considered as violations (or are allowed), and in which conditions. We advocate that these instructions should be considered before signaling a detected TD as an NN violation. In this paper, we are concerned with quantifying how much these instructions influence the results achieved by state-of-theart solutions. We analyzed the public dataset of TDs detected by Glasnost under the regulatory perspective. We found that in specific circumstances, up to 48% of detected TDs cannot be conclusively signaled as NN violations. Our findings point towards the need for additional considerations when designing solutions focusing on NN, and to weaker conclusions drawn by solutions that ignore the regulatory perspective of the Internet.

Type
Publication
2020 IFIP Networking Conference (Networking)