Abordagem para monitoramento da disponibilidade de recursos computacionais utilizando carta de controle I-AM

Abstract

Distributed systems have become fundamental to IT infrastructure, requiring efficient monitoring to ensure the availability and effectiveness of computational resources. Identifying a gap in the literature for effective approaches to monitor CPU availability in such systems, this study proposes an approach that utilizes the I-AM control chart, derived from statistical process control theory, specifically adapted to monitor and manage CPU utilization in distributed environments. The research focused on collecting CPU availability data in simulated distributed systems, analyzing them through I-AM control charts to identify variations in availability and determine optimal levels for resource management. The applied methodology involved a descriptive analysis, employing descriptive statistics, histograms, and time series graphs, to explore the structure, statistical characteristics, and temporal patterns of the collected data. The results demonstrated the effectiveness of the proposed approach in proactively identifying variations in CPU availability, facilitating more robust load balancing and task scheduling strategies. This approach proved not only effective in detecting normal and alert states in CPU availability but also highlighted the importance of parameter calibration to optimize monitoring accuracy. Although the study focused exclusively on CPU availability, neglecting other factors that may influence system performance, the practical implications suggest that the methodology can be extended to other computational resources, thus promoting more efficient and sustainable use of IT infrastructure. It is concluded that the application of statistical quality control techniques in IT environments, especially in distributed systems, offers a promising approach to improve the monitoring and management of computational resources, with significant potential for future research and practical applications in the field.