Zinchenko O. Methodology of building information systems based on virtualization of network functions

As a result of dissertation researches the important scientific and applied problem concerning development of methodology of construction of information systems, on the basis of virtualization of network functions which should function effectively with SDN and NFV is solved. For the first time a mathematical model of an information network with virtualization of network functions was developed, the scientific novelty of which is that it is based on the theory of queuing. This makes it possible to provide real-time self-tuning of network parameters in order to implement the definition of an effective network architecture and increase its productivity. For the first time, a method of reconfiguration of the service chain and the behavior of the SDN control plane was developed, the scientific novelty of which is based on the use of the Mill finite state machine and allows to model the behavior of the SDN control plane. For the first time, a methodology for increasing system performance by predicting the required services and predicting errors has been developed, a scientific novelty of which lies in the fact that it is based on linear programming methods. This enables efficient management of dynamically organized network functions. The method of detecting attacks on the NFV information network has been improved, which, unlike the existing ones, is based on an entropy algorithm and automatically selects countermeasures depending on the type of attack. This makes it possible to determine the streams of packet events that load system resources. To determine the flow of packet events that deplete the controller's resources, the distribution of functions is used by generalizing the entropy of the combined Shannon-Reni formula in real time. For the first time, the concept of building an information system with virtualization of network functions was developed, the scientific novelty of which lies in the fact that it is based on mathematical models of network architecture, forecasting the necessary network services and errors, and methods of parrying network failures at minimal cost. This makes it possible to ensure the timely response of the information system to emergency situations in the event of a network attack, degradation of the structure or deterioration of the functionality of the information system.