Intercomponent interaction in a multi-agent system

The article examines the mechanisms of intercomponent interaction in multi-agent systems. The paper discusses various approaches to messaging between components as well as the advantages and disadvantages of each of them. The key problems of intercomponent interaction are identified and their solutions are proposed. Particular attention is paid to the messaging mechanism based on the message broker. The principles of the broker, its advantages and disadvantages as well as examples of use in multi-agent systems are described. The results of the study showed that the use of the message broker makes it possible to create a flexible and scalable system that can efficiently process a large number of messages and maintain high reliability in operation. The paper presents a description of the data transfer format structure between the components of a multi-agent system. Message routing schemes within the system using a message broker are shown. The configuration for the implementation of the intercomponent interaction schemes is described. A mechanism for encoding messages based on tag keys is proposed, which enables their identification for further processing by software agents. This approach can be useful in the design and development of various multi-agent systems, where it is necessary to exchange messages between different software agents.

1. Yuleisi G.P., Kholod I.I. Interaction in multi-agent data mining systems. Izvestiya SPbGETU LETI = Proceedings of Saint Petersburg Electrotechnical University. 2020;3:18–23. (In Russ.).

2. Skobelev P.O., Ivashchenko A.V., Andreev M.V., Babanin I.O. Multi-agent technologies for managing the distribution of production resources in real time. Mekhanika, upravlenie i informatika. 2011;5:110–122. (In Russ.).

3. Timonin A.N. Overview of the Anylogic simulation tool. Informatsionnye tekhnologii v obrazovanii= Information Technology in Education. 2021;4:231–237. (In Russ.).

4. Liechtenstein V.E., Konyavskii V.A., Ross G.V., Los' V.P. Multi-agent systems: self-organization and development. Moscow, Finansy i Statistica; 2018. 264 p. (In Russ.).

5. Sabir B.E., Youssfi M., Bouttane O., Allali H. Authentication and load balancing scheme based on JSON Token for Multi-Agent Systems. Procedia computer science. 2019;148:562–570. DOI: 10.1016/j.procs.2019.01.029.

6. Gorbunov V.V. Expanded methods of interaction with the service bus of the enterprise. Innovatsii. Nauka. Obrazovanie. 2020;23:152–158. (In Russ.).

7. Ayanoglu E., Aytas Y., Nahum D. Mastering RabbitMQ. Packt Publishing Ltd; 2016. 286 p.

8. Kavalerova A.S., Samochadin A.V., Timofeev D.A. Subscriber authorization module in the rabbitmq queue management system. SPbPU Science Week, 19–24 November 2018, Saint Petersburg. Saint Petersburg, Politekh-Press; 2019. p. 177–180. (In Russ.).

9. Dobbelaere P., Esmaili K.S. Kafka versus RabbitMQ: A comparative study of two industry reference publish/subscribe implementations: Industry Paper. Proceedings of the 11th ACM international conference on distributed and event-based systems, June 2017. 2017. p. 227–238. DOI: 10.1145/3093742.3093908.

10. Vandikas K., Tsiatsis V. Performance evaluation of an IoT platform. 2014 Eighth International Conference on Next Generation Mobile Apps, Services and Technologies. IEEE, Oxford, UK. 2014. p. 141–146. DOI: 10.1109/NGMAST.2014.66.