In this dashlet you will see the objects that will exceed their established maximum limit in the next 30 days.
Because OpMon maintains performance data collection, it uses machine learning algorithms to accurately predict when a given resource exceed your limit.
To do this, the Predictive Analysis dashlet was created in the OpMon Tactical View pane.
Daily the algorithm is executed and makes a predictive analysis of which resources will be exhausted in the next 30 days, based on the behavior of use of this resource in the last 3 months of data collected. The analysis is presented in the Tactical Vision dashboard and updated daily.
The information presented in the RED color indicates the resources that will be exhausted in the next 7 days and the features presented in the YELLOW color indicate the ones that will be exhausted in the next 30 days. It is important to remember that this functionality does not replace the usual monitoring of OpMon, but rather adds important functionality for system administrators to be able to anticipate problems and thus act proactively.
The dashlet also allows the notification in the form of an email or ONC to be configured to allow the sending of the predictive analysis alerts without the need to view the tactical vision panel (to perform this configuration the user must be opmonadmin).
Clicking on the object will take you to the capabilities chart with a predictive analysis based on the data collected from this host/service in the last 3 months. In the capacity chart you can experience different collection and future projection periods.
For Predictive Analysis to work properly, it is very important that all service metrics have their maximum limits properly configured. It is important to note that when the data collected show a weak correlation tendency (Pearson Correlation Coefficient of less than 0.6), the predictive analysis can not be correctly calculated and therefore will not be presented.
The purpose of this functionality is to provide relevant information for predicting future problems, and to treat them before their occurrence.