Intelligent systems based in hospital database malfunction scenarios

Databases are indispensable for everyday tasks in many organizations, particularly in healthcare units. Databases, allows to archive, among other relevant operations, important, private and confidential information about patients clinical status. Therefore, they must be available, reliable and at high perfor- mance level twenty-four hours a day, seven days per week. In many healthcare units, fault tolerant systems are online and ensure the availability, reliability and disaster recovery of data. However, these mechanisms do not allow to take preventive actions in order to avoid fault occurrence. In this context, it is of utmost importance the necessity of developing a fault prevention system. This system can predict database malfunction in advance and provides early decision taken to solve problems. With this paper we inted to monitor the database performance and adapt a forecasting model used in medicine (MEWS) to the database context. Based on mathematical tools it was created a scale that assesses the severity of abnormal situations. In this way, it is possible to define the scenarios where database symptoms must trigger alerts and assistance request