Security Analytics for Protecting Virtualized Infrastructures

Abstract

This paper explains the prevention of attack on the host operating system from the virtual operating system to improve security measures in cloud computing environment. There may be many virtual operating system running on a single host operating system To prevent this, we proposed a security analytics approach using logistic regression that detects advanced attacks using machine learning. Network logs collected periodically generated from the guest virtual machines are stored in the Hadoop Distributed File System. Next, determination of attack presence is performed through logistic regression. Experiments are conducted to evaluate the proposed approach using well-known malware as well as in comparison with existing security techniques for virtualized infrastructure. The results show these detection of malware and network attack using log files generated from virtual operating system Here we use big data concepts and Hadoop file system to store and process large amount of log file generated dynamically.

Keywords: Virtualized operating system, virtualization security, Big Data, security analytics, logistic regression, Hadoop