The recent proliferation of triple-play services has led not only to an increase in the demand for bandwidth for broadband access networks, but also to the need for new Service Delivery Architectures. The goal now is to build service-rich, cost effective and robust environments that would be extremely scalable but also versatile enough to accept the transformations resulting from the introduction of multi-play services. The choice of an appropriate architectural approach and sizing model for the aggregation network is complex and multi-dimensional, encompassing aspects of non-stop delivery, service flexibility, policy management and reduced risk. It is vital to re-engineer the broadband access infrastructure to accommodate intelligent aggregation and optimize for QoS-sensitive services.
To this end, there is a strong need for quantitative models that describe the utility or value of the aggregation network as a function of strategic design characteristics, such as degree of "centralization", single-edge vs. multi-edge characteristics, and degree of clustering, taking crucially into account still not well known characteristics of emerging traffic and usage patterns, geographic distribution of sources and destinations, connectivity of the social networks involved, spatial statistics, all in addition to the temporal characteristics such as degree of self-similarity and burstiness.
Our work is three fold, covering namely, (1) the effect of the characteristics of next generation service traffic and social networking patterns, (2) capturing the basic quantitative design parameters of the network architecture with respect to the degree of centralization and clustering, and (3) taking into account business aspects such as cost models, availability, requirements, customer utility and pricing.