Modeling and performance evaluation of computer systems security operation

D. Guster,N. K. Krivulin
2024-05-22
Abstract:A model of computer system security operation is developed based on the fork-join queueing network formalism. We introduce a security operation performance measure, and show how it may be used to performance evaluation of actual systems.
Cryptography and Security,Systems and Control,Optimization and Control
What problem does this paper attempt to address?
This paper proposes a solution to address the problem of modeling and performance evaluation of computer system security operations. With the rapid development of computer systems and networks, the importance of computer security in organizations is increasing. Research on new security system performance analysis models is crucial due to the potential severe losses caused by ineffective security protections. The authors developed a formal model based on a fork-join queue network to describe the security operations of computer systems. They introduced a security operation performance metric and demonstrated how to use this model to evaluate the performance of real systems. The paper also introduces a security system performance analysis technique and discusses its practical applications. Specifically, the model covers a range of security activities, including attack detection, vulnerability analysis, recovery planning, and software and data recovery. Each operation cycle begins with attack detection, followed by the assessment of system integrity and identification of potential vulnerabilities. Based on the results of the integrity analysis, the software and data recovery process is initiated, and countermeasures against similar attacks are devised. Finally, the operation cycle is concluded by modifying the software, databases, and security policies. The paper also defines the attack cycle time and recovery cycle time and proposes a performance measurement ratio R, which represents the ratio of recovery cycle time to attack cycle time, to measure the proportion of time the system spends in the recovery state. The overall performance of the entire system is evaluated based on the maximum service time, and it is recommended for administrators to prioritize and reduce the time required for the longest process to optimize system performance. Additionally, the model and the related ratio R can be used for effective monitoring of an organization's security system and can be extended to cover security operations processes with different flow and priority constraints. The paper concludes by referencing the formalized approach of fork-join networks and the related matrix theory to describe the dynamics of the model.