Analysis of the influence of the self-similarity factor on the characteristics of mass service systems through quantitative assessment

Problem statement. The latest information technologies have revealed new properties of request flows and procedures in their service, which are reflected in the property of traffic self-similarity. Social networks can be an example of the phenomenon of self-similarity. And the main problem of queuing system is to ensure the QoS. Research Goal. Study the analytical and quantitative assessment of discrepancies in the evaluation of indicators of QS functioning with signs of self-similarity in comparison with the results of their calculations using traditional mathematical models with Poisson flows. Methods. Analyze the theory of QoS, types of distribution, such as Pareto and Weibull, as well as the properties of self-similar traffic. Compare the values of the main parameters that provide the desired quality of service at different inputs and different Hearst parameters. Results. The research focuses on the advantages of using the Weibull distribution over the Pareto distribution when describing phenomena in QS with signs of self-similarity. Conclusions. Research the peculiarities of the influence of the self-similarity property on quantitative QoS parameters in queuing systems.