Paper Review : Enabling Conferencing Applications on the Internet using an Overlay Multicast Architecture


Reviewer : Seh Leng Lim


This paper discusses about the End System Multicast where all multicast functionality is pushed to the edge.


The main contribution of the paper is its detailed performance study of various schemes to construct overlay networks on a wide-area testbed of about twenty hosts distributed around the Internet. 

This is important as End System Multicast has potential to be a serious alternative to IP multicast with its serious scalability and deployment problems for supporting group communication applications over the Internet.


The key main ideas expounded are :

(a) The distinguishable characteristics of conferencing applications of performance requirements, gracefully degradable, session lengths, group characteristics and source transmission patterns

(b) The method of sequential unicast, random, prop-delay-only, latency-only and bandwidth-latency for constructing overlays

(c) It is important to adapt to both bandwidth and latency while constructing overlays optimized for conferencing applications


I think that the paper has a modest contribution (rating of 3) to the study of Internet multicast . Quite a considerable amount of work has previously been done by other authors to the study of overlay construction. Nevertheless, the authors have performed some detailed simulations to support their claims that End System Multicast is realizable in the heterogeneous Internet for performance demanding audio and video conferencing applications . However, as the scale of their  testbed is small, it remains to be seen whether the End System Multicast is really a good multicast implementation for the Internet. As acknowledged by the authors, there is still work to be done to make End System Multicast achieve shorter time scale adaptation in extremely dynamic environments and to study mechanisms for lowering network costs for larger sized groups.


System and application builders may have a better appreciation from this paper of the difficulties involved in implementing video and audio conferencing applications on the Internet.