Performance evaluation of the cloud computing application for IoT-based public transport systems

Abstract

The object of research is cloud computing as an element of the server infrastructure for intelligent public transport systems. Given the increasing complexity and requirements for modern transportation, the application of the Internet of Things concept has a high potential to improve efficiency and passenger comfort. Since the load generated in IoT systems is dynamic and difficult to predict, the use of traditional infrastructure with dedicated servers is suboptimal. This study considers the use of cloud computing as the main server infrastructure for the above systems. The paper investigates the main cloud platforms that can be used to develop such systems and evaluates their advantages and disadvantages. The authors developed the overall architecture of the system and evaluated the performance and scalability of individual components of the server infrastructure. To test the system, a software emulator was developed that simulates the controller module installed in vehicles. Using the developed emulator, stress tests were conducted to analyze and confirm the ability to scale and process input data by the proposed architecture. The test scenarios were developed and conducted on the basis of the existing public transportation system in Kyiv, Ukraine. The experimental results showed that the proposed IoT architecture is able to scale efficiently according to the load generated by the connected devices. It has been found that when the number of incoming messages increases from 40 to 6000, the average message processing time remains unchanged, and the error rate does not increase, which is an indicator of stable system operation. The obtained results can be used in the development of modern public transport systems, as well as for the modernization of existing ones

Effect of the Formation of Hydrophilic and Hydrophobic–Hydrophilic Associates on the Behavior of Copolymers of N-Vinylpyrrolidone with Methyl Acrylate in Aqueous Solutions

“…Further, the question of creating computing systems on non-traditional element bases, including chemical ones, is currently being raised [ 26 , 27 , 28 ]. In particular, in [ 29 , 30 , 31 , 32 , 33 , 34 , 35 ] the question of using stimulus-sensitive polymers to create information-processing systems on a quasi-biological basis was discussed.…”

IoT system development for heart rhythm monitoring and cardiovascular risk estimation

“…The ever-evolving landscape of technological advancements has brought forth the concept of the Internet of Things (IoT), where physical objects and devices seamlessly connect in an open network, capable of self-management, information sharing, and dynamic responses to environmental changes [1][2][3]. Concurrently, IoT is defined as a set of infrastructures facilitating object interconnection, data access, and the application of data mining and analytics processes [4,5].…”

Improving a procedure of load balancing in distributed IoT systems

“…In addition, many studies consider methods for improving the efficiency of load distribution in industrial systems. For example, work [20] investigates the problems of load balancing on the example of a implemented system for monitoring and managing public transport. However, those studies do not fully reveal the problem of load balancing in distributed IoT systems as they only consider available commercial solutions.…”