Intellectual infrastructure for automated control and interoperability of microservices in cloud environments

• Published in: Программные системы и вычислительные методы no. 2; pp. 94 - 114
• Main Authors: Rogov, Dmitriy Vadimovich, Alpatov, Aleksei Nikolaevich
• Format: Journal Article
• Language: English
• Published: 01.02.2025
• ISSN: 2454-0714; 2454-0714
• DOI: 10.7256/2454-0714.2025.2.74760

In the context of rapid growth in the scale and complexity of information systems, the questions of effective integration and support of microservices architectures are becoming increasingly relevant. One of the key challenges is ensuring the interoperability of software components, which implies the ability to reliably exchange data and share information between various services implemented using heterogeneous technologies, protocols, and data formats. In this work, the subject of research is the formalization and construction of an intelligent system ensuring the interoperability of microservice components within cloud infrastructure. A formalized approach is proposed, based on graph, categorical, and algebraic models, which allows for a strict description of data transmission routes, conditions for interface compatibility, and the procedure for automated agreement on interaction formats. An operation for interface agreement is introduced, which identifies the need to use adapters and converters for the integration of various services. Special attention is paid to the task of building a universal interface through which any data streams can be routed, significantly simplifying the process of scaling and refining the microservice system. The developed system architecture encompasses the stages of creation, publication, and deployment of container microservices, automatic verification of data transmission routes, and dynamic management of service states based on load forecasting using artificial intelligence models. The application of the proposed methodology allows for a significant increase in the flexibility, reliability, and scalability of the infrastructure, reduction of operational costs, and automation of the processes of support and integration of new components. The proposed solution is based on a formalized approach to ensuring the interoperability of microservice components within cloud infrastructure. A graph and categorical model is used as a foundation, allowing for a strict definition of data transmission routes and interface agreement procedures between various services. To unify interaction and enhance system flexibility, an interface agreement operation is introduced, as well as the capability for automated identification of the need for data adapters and converters. The developed intelligent load forecasting algorithm allows for dynamic management of component states and rapid adaptation of the infrastructure to changing operating conditions.