ABSTRACT:

A range of business peer-to-peer (P2P) systems for live streaming are introduced lately. The behavior of those well-liked systems has been extensively studied in several measurement papers. Due to the proprietary nature of these business systems, but, these studies should depend on a “black-box” approach, where packet traces are collected from one or a restricted number of measurement points, to infer varied properties of traffic on the management and knowledge planes. Although such studies are helpful to check utterly different systems from the tip-user's perspective, it's robust to intuitively perceive the observed properties while not totally reverse-engineering the underlying systems. In this paper, we have a tendency to have a tendency to explain the network style of Zattoo, one of the largest production live streaming suppliers in Europe at the time of writing, and gift a massive-scale measurement study of Zattoo using info collected by the supplier. To highlight, we tend to tend to found that even when the Zattoo system was heavily loaded with as high as twenty 000 concurrent users on one overlay, the median channel be a half of delay remained less than 2-five s, that, for a majority of users, the streamed signal lags over-the-air broadcast signal by less than 3 s.