NeTS: Small: Continuous Monitoring and Localization of Network Neutrality Violations

Networks that carry application data can apply differential treatment, e.g., by giving different performance to different applications and providers. Such behavior is colloquially referred to as net neutrality violations, and formally as traffic differentiation. For example, in U.S., video streaming is often artificially limited to low data-transfer speeds, meaning that only low-quality video can be supported. Further, this is sometimes done only to certain video-streaming providers, creating an unlevel playing field. This project supports the Wehe software ecosystem, which provides mobile apps and other software that reveals when network providers differentiate service and publishes results for stakeholders. Wehe requires continual maintenance so that Wehe tests continue to correctly identify when they experience differentiation. To address this, this project will bring Wehe software into compliance with current operating systems and apps that experience differentiation and produce strategies to make Wehe more resilient to future changes. Second, Wehe does not isolate which network is responsible for differentiation. To improve Wehe’s ability to identify the correct network responsible for net neutrality violations, this project will develop new statistical methods based on network performance tomography and implement new tests in Wehe that leverage these methods to reliably identify the network responsible for differentiation. The project work will benefit Internet subscribers, policy makers, and service providers by revealing violations of net neutrality, and ensuring that policy decisions are informed by empirical evidence. Internet subscribers can learn whether their providers violate net neutrality via the Wehe mobile app, the general public can see aggregate findings on net neutrality via a public Web site for reporting Wehe results, and findings from the project will be shared with stakeholders such as policy makers. Subscribers will benefit from transparency and understanding alternative options, and content providers can adjust their applications to improve efficiency. The project will also identify unintended negative side-effects of middleboxes, and produce recommendations to improve network efficiency. Details about the project, its findings, and other artifacts are located at: https://wehe.meddle.mobi. The project has produced public datasets, software, results, visualizations, and other insights for six years. The goal is to continue publishing and updating such artifacts in perpetuity, or for as long as the project is relevant. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.