Improvement of Application Monitoring Methods in Microservice Architecture

Abstract

The purpose of this study was to develop approaches to improve the efficiency of monitoring application operations in a microservice architecture by enhancing methods for collecting, processing, and analyzing monitoring data. The authors created an integrated approach that modeled interactions between microservices to identify critical data transmission paths and further optimize architectural components. The proposed mathematical model, based on queuing theory and Markov processes, enabled the assessment of the system’s behaviour under variable load and the prediction of its reliability. One of the crucial steps was the use of machine learning algorithms to automatically detect anomalies in real time, thereby markedly improving monitoring accuracy and reducing false alarms. The development process also incorporated specific requirements for the user interface and system integration. Optimising data collection and processing methods significantly reduced response time to system changes. The study also implemented multi-criteria optimisation methods to effectively manage Service Level Agreement (SLA) and Service Level Objective (SLO) targets, thereby enhancing the system’s adaptability and stability under variable workloads. The application of the developed approach ensures stable system operation, reduced response time, and improved resource management. The research results have significant potential for application in real systems and will contribute to the further development of monitoring technologies in microservice architectures.