Integrating machine learning to the sub-classification of monitoring operations for anomaly detection based on smart automation technology data center.

Abstract Intelligent monitoring of a computer network provides a clear understanding of its behavior at various times and in various situations it also provides relief to support teams who spend most of their time troubleshooting problems caused by hardware or software failures. This type of monitoring ensures the accuracy and efficiency of the network to meets the expectations of its users.However, to ensure intelligent monitoring, it is necessary to start by automating this process which often leads to long and costly interventions. The success of such automations supposes the establishment of predictive maintenance as a prerequisite for good governance of preventive maintenance. However, even when it is practiced effectively, preventive maintenance requires a great deal of time and the mobilization of several full-time resources, especially for large IT structures.This paper gives an overview on the monitoring of a computer network and explains its process and the problems encountered. This article also proposes a method based on machine learning to allow prediction and support decision making to proactively anticipate interventions.