Resumen
The article deals with the processes of data transmission on computer networks in terms of functional and non-functional indicators of network performance. The purpose of the work is to formalize the characteristics of the computer network that are taken into account in the simulation method and to demonstrate the operation of the method using a test case. The following tasks are solved in the article: substantiation of necessity of application of modeling methods during modernization of computer networks; determining the characteristics of a computer network that affect data transmission processes; formalization of indicators that will be directly applied in the modeling process; description of the test case for the model; iterative simulation of network operation. The following research methods are used: basics of system analysis, models of network functioning, simulation modeling method. The following results were obtained: a computer network considered for functional and non-functional performance characteristics, highlighted characteristics that affect the quality of service delivery, and those that affect the cost of the network topology built. Formulas for calculating the amount of information resource of a network are presented. Formulas for calculating the amount of information resource of a network are presented. The list of basic network characteristics that should be taken into account when modeling network load is justified. Test bench for model work is described. An illustrative example of using a test-based simulation method is calculated. Conclusions: It is concluded that taking into account a large number of network performance indicators will overload the modeling process and it is decided to choose universal indicators of a computer network, which would not depend on the topology of its construction or the type of protocol used. The ability to create a simulation model of a computer network for use in predicting network behaviour when changing the number of requests has been confirmed. Further development of the method will allow us to predict the times of network congestion requests to improve the efficiency of the computer network being upgraded.