ABSTRACT
This chapter presents a framework hyperbolic data analytics (HDA) for performing efficient network analytics in large-scale and complex cyber-physical networks. HDA allows computing efficiently and in a scalable manner topological metrics, such as the traffic load centrality and in general distance and path-based metrics, for complex networking topologies, or massive data dependency graphs. The chapter focuses on the efficient computation of the two most important of them, namely routing between’s centrality, which is a path-based metric and its variants, and closeness centrality, which is a distance-based metric. Complex network metrics, such as path-based centrality metrics, the clustering coefficient, become very helpful tools for the analysis and optimization of network structure, while their evolution serves as efficient indicator of the network evolution itself. The authors apply a big data analytics framework, denoted as HDA, which can handle the required computation of network metrics in large-scale environments efficiently.
