A New Approach for Selecting Aggregated Multicast Trees to Reduce Forwarding States

Multicast is an efficient method for sending data to multiple destinations in a single transmission. Conventional multicast routing often suffers from scalability issues. Here, the number of forwarding states maintained in the Network layer generally increases with the number of concurrently active multicast groups. Consequently, large-size routing tables get generated, which in turn causes degradation in router performance. There exist only a limited number of studies on aggregated multicast to address this issue. The existing studies such as AM (Aggregated Multicast), STA (Scalable Tree Aggregation for Multicast), and STS (Shared-Tree Selection for Aggregated Multicast) attempt to perform tree aggregation, however, still retain high number of forwarding states in the routers. Besides, they mostly ignore the effect of tree aggregation ratio in terms of network performance metrics (for example delay and throughput). Hence, the research on aggregated multicast is still at an elementary stage. To this extent, in this study, we propose a novel aggregated multicast approach to reduce Network layer forwarding states. Our approach proposes new methods for selecting aggregated multicast trees, refining search range, and replacing stale trees in process of the aggregation. We show the effect of our proposed tree aggregation in terms of network performance metrics (delay and throughput). We also show that performance of our approach is better than previous approaches by comparing different performance metrics such as the number of trees, the number of forwarding states, delay, and throughput. Ns-3 simulation results confirm that our approach can reduce up to 92% forwarding states and 34% delay compared to conventional multicast, and reduce up to 88% forwarding states and 29% delay compared to the STS method.