IMAM: uma ferramenta para monitoramento de sistemas e dispositivos em infraestruturas críticas de IoT baseada em Aprendizado de Máquina

Abstract

Faults in critical systems and devices should be dealt with quickly and efficiently. Inactivity periods can be costly and have significant consequences in several contexts. It is essential that information systems are always available and reliable. Although most infrastructure monitoring tools can identify faults, above all, is important to obtain knowledge from infrastructure data in several situations, including failures and, especially, circumstances that precede such flaws. These infrastructure’s knowledge becomes much more important, as it is desired to predict possible anomalous behaviors from systems and devices monitoring log data and to support actions to ensure availability and fault tolerance proactively. Aiming to address these challenges, this work presents IMAM, a tool capable of monitoring systems’ availability and collecting, storing and analyzing IoT-based critical infrastructure monitoring logs through Machine Learning techniques.