Modeling Methods Complex Computer Systems

The paper discusses the methods of modeling computer systems that are applicable for developing an algorithm that allows to evaluate the current state of the observed object in terms of information security, to identify unsafe states of a real system by interpreting the values of the observed parameters. The authors describe the existing problems of modern automated information security systems, indicate the shortcomings and limitations that lead to low efficiency and high cost of ownership of such systems. The paper provides an overview of the methods that can be used to develop an algorithm for analyzing the parameters of an information system that recognizes whether the current state of a real system belongs to one of the classes (“safe” or “unsafe”). In particular, the following systems modeling methods are given – the reduction method, based on multi-agent systems, and modeling based on agents. The reduction method is considered on the example of cluster analysis. For multidimensional data, the use of Kohonen self-organizing maps is most appropriate. For complex modeling methods (based on multi-agent systems and agent-based), their basic concepts (agents and their properties, types), a brief description of the methods are introduced.