Viewing YouTube from a metropolitan area: What do users accessing from residential ISPs experience?

In this work, we perform a controlled study on the perceived experience of viewing YouTube videos as observed from the end users' point of view through their residential ISPs in a metropolitan area. This study is conducted using our tool Pytomo, which we developed to emulate the end users' experience of viewing YouTube videos. Pytomo crawls and downloads YouTube videos to collect a number of measures, including information about the YouTube servers that are delivering them. This open-source tool was provided to a group of volunteers located in the Kansas City metropolitan area. These volunteers, who use different residential ISPs to access the Internet, were instructed to synchronously run the tool to collect the measurement data. Based on the data collected over specific time windows (separated by three months), we observed that there is a noticeable difference in the quality of experience depending on the residential ISPs. Furthermore, the content distribution policies for YouTube, for different residential ISPs vary and the round trip time is not the primary factor for choosing video servers.