Modern video conferences are increasingly moving to the WebRTC technology, which works in a browser. Almost all modern video conferencing systems provide functionality for maintenance, transcoding, and recording of several video streams. The proposed platform provides the opportunity to teach students using video conferences and integrates into Moodle. The goal of this research is to study its operation based on Janus - the open source WebRTC server.
In some areas neural networks are becoming a non-alternative way of solving problems. Recognizing images, sounds, classification - these problems require serious processor power and memory for neural network learning and operation. Modern mobile devices have quite good characteristics for executing only the primary layers of deep neural networks, but there are not enough resources for entire networks. Since the training of non-mobile networks for mobile devices takes place separately on external resources, a method was developed for the distributed operation of a neural network with vertical partitioning sets of layers with synchronization of learning data. In some cases, the proposed approach allows to use full-sized deep neural networks on mobile device, where they are needed without overloading the communication channel and device resources.
This research is aimed at studying the features of the simplified implementation of the protocol stack, the efficiency of routing and performance in the transit mode of devices data transfer with priority for the longest battery life.
1 January 2021 in Vestnik of Ryazan State Radio Engineering University
This paper studies the features of simplified implementation of the protocol stack for ad-hoc Wi-Fi networks. We describe custom routing protocol and study its efficiency and productivity in the transit data connection mode. The latency and throughput of this way of routing are within acceptable ranges. This routing method can be used even for unstable communications with gateways. This is important if the network has battery-powered devices, which work periodically.
Modern trends in the information technology have led to the fact that entire systems of infrastructure are becoming software-defined. Modern hyper-converged solutions use software-defined networking and soft switches for the hypervisor networking subsystem. The paper goal is to study traffic processing in hyperconverged structures with software switching based on OpenFlow versus traditional approaches. The features of the hyperconverged solutions network infrastructure are considered, approaches to the study of software-defined environments are described. A model of the processing traffic internal structure of a converged node, combining the functions of a hypervisor, a storage system and a switch, is proposed. Interface models reproduced traffic switching with the traditional approach and with higher-level OpenFlow processing have been developed. The approaches to the implementation of the developed models based on experimental studies of network equipment are described. The results of an experimental study of a network node and a synthesized model are presented. The possibility of implementing the proposed approaches within the specified accuracy are described.