Investigation of Information Flow Properties in the Main Internet Channel
Students Name: Vakhtina Mariia Vitaliivna
Qualification Level: magister
Speciality: Information Communication Networks
Institute: Institute of Telecommunications, Radioelectronics and Electronic Engineering
Mode of Study: part
Academic Year: 2023-2024 н.р.
Language of Defence: ukrainian
Abstract: Today, computer networks (CNs) are widely used to transmit various information, monitor and control various services in real time, watch TV programmes, make online purchases, etc. Due to the increase in new classes of telecommunication devices and corresponding services, the amount of information transmitted via the Internet is growing rapidly [1,2]. At the same time, the actual structure of the transmitted information created and used by numerous users of personal computers, smartphones, tablets, TVs, household appliances, etc. has become significantly more complex. In this situation, the requirements for flexibility and scalability of modern CNs are increasing, the properties of which are significantly different from those of CNs with a classical architecture (essentially static) [3,4]. For example, traditional CNs architectures/designs are ineffective in dynamic environments. At the same time, classical approaches focused on distributed management of traditional CNs devices (e.g., virtual networks (VLANs)) do not meet the current level of development of server and storage virtualisation, as well as the requirements of large businesses and service providers. In its 2016 analytical report, KPMG introduced the concept of "disruptive traffic", i.e. traffic of such volumes that the network equipment in use cannot handle and "disruptive technologies" that create disruptive traffic (virtual reality, cloud services, artificial intelligence, real-time data analysis, etc.) To effectively address the problems of "disruptive traffic", designing equipment that meets the needs of modern CMs, as well as designing a new generation of CMs, including virtual software-defined networks SDN, it is necessary to understand the features of information flows transmitted in modern CMs, the mechanisms of their interaction with each other and the impact on channel utilisation. Internet traffic is a complex multidimensional object that can be studied in different dimensions (e.g., the number and size of transmitted packets, the amount of information transmitted, etc.) [5]. The first and integral stage of quantitative analysis of dump data is the stage of semantic analysis (parsing) of Internet traffic dumps (pcap files), where the necessary quantitative information is obtained from these files, which is then used to obtain quantitative estimates of traffic characteristics. Thus, it is relevant to study the properties of information flows in high-speed trunk Internet channels based on a systematic approach.