DYNAMIC RESILIENCE MECHANISM FOR SCALABLE INFORMATION INFRASTRUCTURES

Abstract

This article examines the challenge of ensuring the stability of computing systems under peak loads. The main problem is that traditional monitoring methods often fail to detect critical failures and unforeseen situations on time, which may lead to data loss and significant economic damage. The research aims to develop a hybrid monitoring architecture that combines synchronous polling of critical parameters with asynchronous event-driven data collection. To achieve this goal, the proposed solution employs virtual probes that allow for system analysis without significantly impacting performance and a component capable of detecting anomalous system states.Experimental studies have confirmed the effectiveness of the proposed model, which reduces the risk of failures, optimizes the use of computing resources, and ensures high system scalability under various load conditions. The proposed model was tested on real-world scenarios using simulation environments that emulate emergency situations and intensive query streams. The research results demonstrate a significant improvement in system efficiency, a reduction in response time to failures, and an optimization of information systems’ performance. The obtained data also indicates the possibility of integrating the proposed approach with existing solutions for monitoring and managing computing systems.Considerable attention was paid to analyzing the impact of various monitoring parameters on performance, which allowed the determination of optimal values for balancing data collection accuracy and system load. A comparative analysis with traditional monitoring methods was also conducted, revealing that the hybrid approach provides a more stable system operation during peak loads. The summarized research findings may serve as a foundation for further scientific investigations and implementing innovative technologies in computing resource management.