Visual Analytics for Root DNS Data

摘要: The analysis of vast amounts network data for monitoring and safeguarding a core pillar the internet, root DNS, is an enormous challenge. Understanding distribution queries received by how those change over time, in intuitive manner sought. Traditional query performed packet packet, lacking global, temporal, visual coherence, obscuring latent trends clusters. Our approach leverages pattern recognition computational power deep learning with 2D 3D rendering techniques quick easy interpretation interaction amount DNS traffic. Working real-world experts, our visualization reveals several surprising clusters queries, potentially malicious benign, discovers previously unknown characteristics DDOS attack, uncovers unforeseen changes time. These discoveries will provide analysts deeper understanding nature traffic under their charge, which help them safeguard against future attack.