Read This Before You Begin

Here are some basic concepts regarding exploring data and building dashboards.

A data explorer, in the context of Oracle Cloud services, is an interface that lets you explore collected IT operational data, perform forecasting, view trends, customize the display of this data, and save it for future use. In this case, views of saved searches are called widgets and a collection of custom widgets can form a dashboard.

A data explorer target is the entity whose data is being collected, for example the Oracle WebLogic Server. Targets can be of various types, for example the Oracle WebLogic Server is a Middleware type of target.

Data collected from various targets is in the form of metrics.

Heat maps (also known as treemaps) are graphical representations of data displayed as a combination of colored rectangular cells, each representing an element of data. Heat maps help visualize the state and effect of a large number of elements at once.

Machine learning algorithms are constantly looking at the data collected by Oracle Management Cloud. There are four major classes of machine learning algorithms:

• Anomaly detection is the capability to study behavior-specific metrics over time.

• Clustering is applied to unstructured data such as logs, to identify patterns, and then match the data to that pattern to quickly identify trends and outliers in logged data.

• Prediction is the capability to forecast future values of time-series metrics based on historical data and system behavior.

• Correlation is the ability to determine how sets of metrics correlate in behavior.