Dmytrenko V. Model and method of building convergent software-configured networks with guaranteed reliability of control information.

ANNOTATION Dmytrenko V.V. Model and method of building convergent software-configured networks with guaranteed reliability of control information. – Qualifying scientific work on manuscript rights. Dissertation for obtaining the scientific degree of Doctor of Philosophy in the specialty 123 "Computer Engineering" – State University of Information and Communication Technologies of the Ministry of Education and Science of Ukraine, Kyiv, 2024. The dissertation is devoted to the development of a model and method of building a simulation-analytical model of convergent software-configured networks for the study of packet delay under various load characteristics of traffic processing processes with guaranteed reliability of control information. Solving the task of minimizing SDN network packets directly depends on the quality parameters of converged software-configured networks. The resulting models generated by the software and hardware complex can be used to build a network and study its performance parameters. The developed methods of creating models of input and output parameters of convergent software-configured networks allow to increase the quality indicators of data transfer, services and networks, as well as the guaranteed reliability of control information. Modern high-speed data transmission networks must have predictable behavior for any combination of traffic parameters and equipment settings. One of the main devices of high-speed highways are switches, which gradually became more complicated, increasing functionality. Today, ISPs are beginning to use software-configured networks that provide more flexible approaches to traffic management. Software-configured networks operating on the basis of the OpenFlow protocol are such an approach to the construction of information communication networks, in which network management functions are separated from the level of data transmission and transferred to a special server-controller. They significantly improved network manageability and allowed to combine virtual and physical infrastructure under a single guaranteed management. To date, simulation modeling is the main tool for SDN network research. A fairly large number of network simulation systems have been developed, relying in their work on such parameters as the number of nodes, connections between them, protocols, characteristics of the data transmission channel, as well as performed functions. The probabilistic time characteristics of telecommunication network traffic processing depend on the time of operation of all processing traffic processes, including the operation of queues, filters, settings, routing and switching, as well as on resource loading. The processing time of the same package on the same equipment with different settings can vary by a factor of ten or more. Traffic processing steps that are not considered in standard models add additional delays. The most common network is a converged software-configurable network, which combines the implementation of basic functions in both traditional and software-configurable mode. The dissertation analyzes approaches to automating the transfer of characteristics from the network to the modeling system. This made it possible to significantly simplify the process of building models of specific devices and networks, as well as automatically obtain the characteristics of specific models of equipment and control the reliability of models using statistical methods. The method of building a convergent simulation-analytical model of a telecommunications network with guaranteed reliability of control information is used for the construction of information communication networks of Ukraine.