An utility-driven routing scheme for scaling multicast applications

Abstract

Multicast is a common platform for supporting group communication applications, such as IPTV, multimedia content delivery, and location-based advertisements. Distributed hash table (DHT) based overlay networks such as Chord and CAN presents a popular distributed computing architecture for multicast applications. However, existing research efforts have been mostly dedicated to efficient message delivery techniques to alleviate the influence of network dynamics on geo-distance based routing, such as reducing the delivery path length or optimizing routing path by utilizing network locality. In this paper, we argue that the geo-distance based routing protocols used in existing overlay networks are inefficient in terms of both resource use and environmental accommodation for multicast applications. We devise a utility driven routing scheme to improve the routing efficiency with three unique features. First, our utility function is defined based on a careful combination of hop counts and routing path latency. Second, we use CAN-like routing as an example and extend it by utilizing shortcuts to reduce the routing path length and by introducing a utility function to combine path latency with geo-distance based metric in determining the near-optimal route for each routing request. Third and most importantly, our utility function is designed by using a tunable influence parameter to allow nodes to adaptively make the most promising routing decision according to their specific network state and circumstances, such as overlay connectivity and next hop latency. Our experimental evaluation shows that the utility-driven routing scheme is highly scalable and efficient compared to existing geo-distance based routing protocols and demonstrates that by combining shortcuts, path latency with geo-distance can effectively enhance the multicast delivery efficiency for large scale group communication applications.