A Security Visualization Analysis Methodology for Improving Network Intrusion Detection Efficiency

The flood of raw data generated by intrusion detection and other network monitoring devices can be so overwhelming that it causes great difficulty in detecting patterns that might indicate malicious traffic. In order to more effectively monitor and process network and forensic data within a virtualized environment, Security Visualization (SecViz) provides software-based visual interfaces to analyze live and logged network data within the domains of network security, network and cloud forensics, attack prevention, compliance management, wireless security, secure coding, and penetration testing. Modern networks generate enormous amounts of data that is often stored in logs. Due to the lack of effective approaches to organizing and visualizing log data, most network monitoring tools focus at a high level on data throughput and efficiency, or dig too far down into the packet level to allow for useful analysis by network administrators. SecViz offers a simpler and more effective approach to analyzing the massive amounts of log data generated on a regular basis. Graphical representations make it possible to identify and detect malicious activity, and spot general trends and relationships among individual data points. The human brain can rapidly process visual information in a detailed and meaningful manner. By converting network security and forensic data into a human-readable picture, SecViz can address and solve complex data analysis challenges and significantly increase the efficiency by which data is processed by information security professionals. This study utilizes the Snort intrusion detection system and SecViz tools to monitor and analyze various attack scenarios in a virtualized cloud computing environment. Real-time attacks are conducted in order to generate traffic and log data that can then be re-played in a number of software applications for analysis. A Java-based program is written to aggregate and display Snort data, and then incorporated into a custom Linux-based software environment along with select open-source SecViz tools. A methodology is developed to correlate Snort intrusion alerts with log data in order to create a visual picture that can significantly enhance the identification of malicious network activity and discrimination from normal traffic within a virtualized cloud-based network