Minimal bandwidth multicast-broadcast routing algorithms

The purpose of this paper is to reduce the communication costs for applications that send the same data to multiple recipients across a wide-area network. The existing multicast and broadcast routing mechanisms are not efficient in terms of storage, bandwidth consumption and traffic concentration: data packets or membership information are occasionally sent over many links that do not lead to the receivers or senders, respectively. This paper constructs two different delivery trees, the collection of nodes (routers) and the links that a multicast or a broadcast packet traverses, by two heuristic algorithms. The first tree is a single spanning tree to deliver broadcast packets with minimal network cost and minimal number of traversed links. The second tree is a single minimal bandwidth group-shared tree (MBGST) per group, to deliver a multicast packet. This tree is a construction of shortest-path multicast tree (SPT) in order to minimize the number of traversed links. The results indicate a 45% reduction of the multicast load independent from the network growth. This realizes a greater traffic concentration and it balances the load over the traversed routers (nearly equal fan-out). This reduces the maximum fan-out of each node, which avoids the CLR increasing at the switching node. The multicast MBGST tree is suitable for constructing the group-shared tree required by PIM-SM protocol, which uses both SPT and group-shared tree for multicast packet delivery according to the application´s delay sensitivity. The correctness, stability, optimality, fairness in distributing load over the traversed routers, and scaling properties of these algorithms make it well suited for large heterogeneous networks.