Srautinių atakų įtakos internėtinės paslaugos sutrikdymui modeliavimas ir tyrimas

In the dissertation the Denial of Service (DoS) attacks and their models are investigated. DoS attack is a type of cyber attacks when an attacker tries to deny a service in the network machine. There are many types of DoS attacks, and therefore the main object of the dissertation is specified as distributed denial of service (DDoS) attacks. DDoS uses multiple agents at the same time to exhaust certain resources of network machine and make it unavailable. The importance of DDoS attacks can be explained on the basis of the following facts: nowadays there are no countermeasures which can ensure full resistance to DDoS; DoS effect can be created even by legitimate users of the systems; internet services become more popular therefore the denial of such a service or diminishing of its quality can cause undesired impact on the other systems or their users. The main objective of this dissertation is creation of model for the estimation of the composite DDoS attack success. This model would allow estimating of network machine resistance to different type and power DDoS attacks. The dissertation consists of eight parts including Introduction, 5 chapters, Conclusions and References. In the introduction, the investigated problem, importance of the thesis and the object of research are defined and the purpose and tasks of the thesis, scientific novelty are described together with the practical significance of results and defended statements. At the end of introduction, author’s publications on the subject of the thesis are listed. They introduce and reflect the material of presentations at the conferences and justify the structure of the dissertation: Chapter 1 – revision of DoS attacks, analysis of existing DoS attack and their countermeasure taxonomies. In this chapter, review of the specificities of the computer network modelling and analysis of the existing DoS and DDoS attack models is presented as well. Chapter 2 – introduction and description of the new taxonomies for DoS attack and their countermeasure. Chapter 3 – description of proposed models for different types of DDoS attacks as well as composite DDoS attack model. Chapter 4 and 5 – investigation of the results of proposed models. They are compared with the results of the OPNET modelling tool, and different situations are analysed to represent characteristics of the DDoS attacks. Conclusion is made that the proposed composite DDoS attack model is easily adaptable for different types of DDoS attacks and might be used to define probability of attack success