A Data-Driven Analysis and Tuning of a Live Hybrid CDN/V2V Video Distribution System

Video live streaming now represents over 34.97% of the Internet traffic. Typical distribution architectures for this type of service heavily rely on CDNs that enable to meet the stringent QoS requirements of live video applications. As CDN-based solutions are costly to operate, a number of solutions that complement CDN servers with WebRTC have emerged. WebRTC enables direct communications between browsers (viewers). The key idea is to enable viewer to viewer (V2V) video chunks exchanges as far as possible and revert to the CDN servers only if the video chunk has not been received before the timeout. In this work, we present the study we performed on an operational hybrid live video system. Relying on the per exchange statistics that the platform collects, we first present an high level overview of the performance of the system in the wild. A key performance indicator is the fraction of V2V traffic of the system. We demonstrate that the overall performance is driven by a small fraction of users. By further profiling individual clients upload and download performance, we demonstrate that the clients responsible for the chunk losses, i.e. chunks that are not fully uploaded before the deadline, have a poor uplink access. We devised a work-round strategy, where each client evaluates its uplink capacity and refrains from sending to other clients if its past performance is too low. We assess the effectiveness of the approach on the Grid5000 testbed and present live results that confirm the good results achieved in a controlled environment. We are indeed able to reduce the chunk loss rate by almost a factor of two with a negligible impact on the amount of V2V traffic .