Scalable network intrusion detection and countermeasure selection in virtual network systems

Abstract

Security of virtual network systems, such as Cloud computing systems, is important to users and administrators. One of the major issues with Cloud security is detecting intrusions to provide time-efficient and cost-effective countermeasures. Cyber-attacks involve series of exploiting vulnerabilities in virtual machines, which could potentially cause a loss of credentials and disrupt services (e.g., privilege escalation attacks). Intrusion detection and countermeasure selection mechanisms are proposed to address the aforementioned issues, but existing solutions with traditional security models (e.g., Attack Graphs (AG)) do not scale well with a large number of hosts in the Cloud systems. Consequently, the model cannot provide a security solution in practical time. To address this problem, we incorporate a scalable security model named Hierarchical Attack Representation Model (HARM) in place of the AG to improve the scalability. By doing so, we can provide a security solution within a reasonable timeframe to mitigate cyber attacks. Further, we show the equivalent security analysis using the HARM and the AG, as well as to demonstrate how to transform the existing AG to the HARM.