Modeling, identifying, and emulating dynamic adaptive streaming over HTTP
Andrew C. Reed · Carolina Digital Repository (University of North Carolina at Chapel Hill) · 2019
As HTTP-based streaming video applications have grown to become a major source of Internet traffic, and as the new ISO standard Dynamic Adaptive Streaming over HTTP (DASH) gains industry acceptance, researchers need the ability to both (i) study real-world viewing data and (ii) emulate realistic DASH streams in network experiments. The first effort is complicated by the fact that researchers are often restricted to anonymized, header-only (i.e. payload-truncated) traces. The second effort is difficult since the process of encoding videos for DASH results in numerous large files and since popular videos are subject to restrictive copyright law. In this thesis we present our work towards developing a model for DASH traffic and show how the model can be applied to identify specific DASH videos in anonymized, header-only traces. We also present our solution for emulating DASH using compact representations of both DASH services (e.g. Netflix and Amazon) and videos.