Multistage anomaly detection over intelligence threats for secured cloud computing resources

Abstract

Many research institutions (ENISA, CSA, SUN, etc.) on cloud computing are always referred to a DDoS attacks item warningly. In this paper, we describe the evolution of intelligence threats to APTs from various DDoS attacks, symptoms of DDoS attacks, and multistage anomaly detection scheme of anomaly for secured cloud computing resource. Specially, Lightweight anomaly detection stage could classified volume data into large volume data and small volume data, and applied Bayesian techniques to detect anomaly and symptoms of various attack. And focused anomaly detection stage is performed to detect novel attacks by unsupervised learning algorithm. © 2013 International Information Institute.