Gossip-Based Video Streaming: Managing Heterogeneous Bandwidth

Gossip-based information dissemination protocols are considered easy to deploy, scalable and resilient to network dynamics. Load-balancing is inherent in these protocols as the dissemination work is evenly spread among all nodes. Yet, large-scale distributed systems are usually heterogeneous with respect to network capabilities such as bandwidth. As such, gossip's blind load balancing may significantly hamper the performance of the data dissemination. This talk presents HEAP, an  HEterogeneity-Aware gossip Protocol, where nodes dynamically adapt their contribution to the gossip dissemination according to their bandwidth capabilities. HEAP dynamically leverages the most capable nodes by increasing their fanouts, while decreasing by the same proportion those of less capable nodes. This preserves the simple and proactive nature of gossip, while significantly improving its effectiveness. Experiments in the context of a video streaming application on a PlanetLab testbed show that HEAP significantly improves the perceived quality over standard homogeneous gossip protocols, especially when the stream rate is close to the average available bandwidth.