A Locating-First Approach for Scalable Overlay Multicast

TL;DR

This paper introduces a scalable locating algorithm for overlay multicast that efficiently directs new nodes to their closest peers, enabling a robust, topology-aware overlay with low overhead suitable for large-scale applications.

Contribution

It presents a novel locating-first approach and a scalable clustered hierarchical overlay scheme called LCC, improving scalability and efficiency over existing topology-aware multicast methods.

Findings

Locating process uses modest resources in time and bandwidth.

LCC outperforms existing schemes in scalability and robustness.

Experimental results on PlanetLab confirm effectiveness for large-scale multicast.

Abstract

Recent proposals in multicast overlay construction have demonstrated the importance of exploiting underlying network topology. However, these topology-aware proposals often rely on incremental and periodic refinements to improve the system performance. These approaches are therefore neither scalable, as they induce high communication cost due to refinement overhead, nor efficient because long convergence time is necessary to obtain a stabilized structure. In this paper, we propose a highly scalable locating algorithm that gradually directs newcomers to their a set of their closest nodes without inducing high overhead. On the basis of this locating process, we build a robust and scalable topology-aware clustered hierarchical overlay scheme, called LCC. We conducted both simulations and PlanetLab experiments to evaluate the performance of LCC. Results show that the locating process entails modest resources in terms of time and bandwidth. Moreover, LCC demonstrates promising performance to support large scale multicast applications.