We describe a new scalable application-layer multicast protocol, specifically designed for low-bandwidth, data streaming applications with large receiver sets. Our scheme is based upon a hierarchical clustering of the application-layer multicast peers and can support a number of different data delivery trees with specific desirable properties. We show that group members maintain state for a constant number of other membersand the control overhead is also a constant.

Due to the sparse deployment of IP multicast in the Internet today, some researchers have proposed application layer multicast as a new approach to implement widearea multicast services. In this approach multicast functionality is implemented at the end-hosts instead of network routers. Unlike network-layer multicast, application layer multicast requires no infrastructure support and can be easily deployed in the Internet. In this paper, we describe a set of application layer multicast protocols that have been proposed in recent literature, classify them based on some properties and present a comparison of performance and applicability of these schemes.

Application layer multicast protocols organize a set of hosts into an overlay tree for data delivery. Each host on the overlay peers with a subset of other hosts. Since application layer multicast relies only on an underlying unicast architecture, multiple copies of the same packet can be carried by a single physical link or node on the overlay. The stress at a link or node is defined as the number of identical copies of a packet carried by that link or node. Stretch is another important metric in application layer multicast, which measures the relative increase in delay incurred by the overlay path between pairs of members with respect to the direct unicast path. In this paper we study the NICE application layer multicast protocol to quantify and study the tradeoff between these two important metrics --- stress and stretch in scalably building application layer multicast paths.