Seminar

Speaker: Ivan Tatarinov

Research on network traffic is an important stage in ensuring the security and optimizing the performance of computer networks. Analyzing various characteristics of network traffic, including the probability distribution of data transmission speed, understanding the behavior patterns of net-work traffic, and detecting anomalies, such as those resulting from DDoS attacks, are key aspects of this type of research. This paper describes the development and utilization of a toolkit that pro-vides a full range of tasks for analyzing network traffic, from collecting data from various types of computer networks to visualizing the analysis results.
The massive volumes (in the case of researching network traffic in large backbone channels with millions of network devices) and the diversity of network traffic (in the case of studying net-work traffic in small computer networks, including home computer networks) require both high performance and flexible management with fine tuning, which are often not satisfied by standard tools. In the context of network traffic research, the necessity arose to develop a custom toolkit that is flexible and powerful enough for effectively analyzing large-scale network data.
To carry out tasks related to the study of network traffic, a comprehensive set of tools was developed, covering all stages of network traffic analysis, including data interception, storage, di-rect analysis of stored data, and visualization of results. Each component of the toolkit is optimized to provide high efficiency and accuracy in processing and interpreting network data.
Each stage of network traffic analysis required the application of specific approaches to achieve the set goals. For example, during the data interception and storage stage, post-processing was required to significantly reduce both the volume of stored data and the speed of subsequent data retrieval. During the network traffic analysis stage, algorithms for rapidly calculating parame-ters for approximating curves and so on were required.

Сonnecting to Webinar.
Information on the seminar and the link to connect are available at Indico.