ANOVA Simultaneous Component Analysis for the Efficient Exploration of Massive Network Traffic

The notion of network observability has been behind network monitoring and management practices from the early times. In the era of Big Data, observability is not only a matter of devising the best data measurement techniques (e.g., the network telemetry framework in RFC9232), but also of properly engineering good practices for data visualization, exploration, and understanding. In this paper, we extend ANOVA Simultaneous Component Analysis (ASCA), for the visualization of Big Data collected from a network. ASCA is a combination of Analysis of Variance (ANOVA) and Principal Component Analysis (PCA) employed mainly in the clinical sciences. With this Big Data extension of ASCA, we provide insights into the massive and complex data of the Netmob 2023 Data Challenge.