Ing. Daniel Sedlák

Článek

10.1007/s10586-025-05251-x

Katedra počítačových systémů

In many clouds running around the world, much of the communication takes place over TCP. Inside clouds, a huge number of new TCP connections are created every second, carrying huge amounts of information. Different variants of TCP use different algorithms to control data flow, which vary in complexity and suitability for specific conditions. Our team has experimentally examined 17 commonly used TCP variants in detail, resp. measured their achievable throughput and energy consumption in a data center network environment for different lengths of duration. Based on the measurements, we calculated the energy efficiency of all TCP versions and compared it among them. We analyzed the network traffic in a real DC and determined the possible improvement based on the facts we gained. Adopting the recommendations from our conclusion can help to reduce the power consumption in east–west cloud communication by units or tens of percent at a global scale simply by ensuring suitable TCP variants.

Stať ve sborníku

10.1109/IC2E61754.2024.00039

Katedra počítačových systémů

A key responsibility for Data center (DC) operators is monitoring, which involves analyzing logs, aggregating traffic statistics, and assessing hardware utilization. Monitoring is vital for troubleshooting and predicting traffic patterns. This paper introduces a new highly scalable, easy configurable framework for utilizing sFlow technology to collect and process network statistics of DC network communication, like connection durations, detection of traffic anomalies, etc. We tested the proposed framework in production, thanks to a partnership with a privately owned DC that provides Infrastructure as a Service worldwide.

Stať ve sborníku

10.1109/CloudNet62863.2024.10815853

Katedra počítačových systémů

Data centers create the backbone of the modern Internet. However, internal network traffic characteristics are closed know-how of data centers. We have collected an internal network traffic analysis based on the data from one of major world data centers. We have analysed 4 internal network traffic characteristics (clustering of IP addresses, application traffic patterns, frequency of changes in cluster topologies and histograms of communicating IP address pairs). The data set has been published at GitHub.

Článek

10.1016/j.dib.2023.109335

Katedra počítačových systémů

Most of the video content on the Internet today is distributed through online streaming platforms. To ensure user privacy, data transmissions are often encrypted using cryptographic protocols. In previous research, we first experimentally validated the idea that the amount of transmitted data belonging to a particular video stream is not constant over time or that it changes periodically and forms a specific fingerprint. Based on the knowledge of the fingerprint of a specific video stream, this video stream can be subsequently identified. Over several months of intensive work, our team has created a large dataset containing a large number of video streams that were captured by network traffic probes during their playback by end users. The video streams were deliberately chosen to fall thematically into pre-selected categories. We selected two primary platforms for streaming - PeerTube and YouTube The first platform was chosen because of the possibility of modifying any streaming parameters, while the second one was chosen because it is used by many people worldwide. Our dataset can be used to create and train machine learning models or heuristic algorithms, allowing encrypted video stream identification according to their content resp. type category or specifically.