A content-based publish/subscribe framework for large-scale content delivery

The publish/subscribe communication paradigm has become an important architectural style for designing distributed systems and has recently been considered one of the most promising future network architectures that solves many challenges of content delivery in the current Internet. This work is concerned with scaling decentralized content-based publish/subscribe (CBPS) networks for large-scale content distribution. A fundamental step for CBPS networks to reach the large-scale is to move from the current exhaustive filtering service model, where a subscription selects every relevant publication, to a service model capturing the quantitative and qualitative heterogeneity of information consumers requirements. Moreover, the proposed work aims at leveraging caching for increasing the communication efficiency of CBPS operating at large-scale characterized by widely spread information consumers with heterogeneous requirements, large number of publications and scarcity of end-to-end bandwidth. We propose and design a service model for addressing the consumers’ requirements for content-based information retrieval and describe the relevant protocols necessary to implement such a service. We evaluate the proposed approach, by using realistic workload scenarios and comparing different content and interest forwarding strategies as well as caching policies in terms of resource efficiency and user perceived QoS metrics.