Hi Bruno,

Regarding your first question, this one can be implemented using the existing data sources:

1. Data Source: Parameters
   Note: If you select 'Protocol', you will need to add an extra filter that will contain the element where the column parameter (that you would like to display in the trend line graph) is located

   Once you select the data source, you will get the list of parameters. From the list of parameters select the column parameter that you would like to display in the trend line graph.

2. Filter: Indices
   Select the index from the table where the column parameter is located

Below an example using the connector 'Microsoft Platform':

Regarding the second question, a first approach will be to use the aggregation rules available on views. Once you create the aggregation rule, you can see as data source in dashboard:

1. Data Source: Parameters
   In this case, select 'Views'. Once you select the view you can select the aggregation rule that you would like to display
2. Filter: Not required

The main drawback for this approach is that you will need to create a view that contain a single element.

Another approach is more complex and requires the DxM module DataAggregator:

• Create a query that will contain the aggregated value(s).
• Using the DataAggregator, execute the query at the desired frequency. The DataAggregator will generate a session record (e.g. CSV file) that contain the result of the query created in the previous step
• For data retention, these session records can be offloaded to an external database (e.g. Elasticsearch). This can be done for example through a connector that implements a logger table. The connector should take care of offloading the session records in the logger table at a specific frequency. The logger table should contain at least the following columns:
  • Timestamp: This column will be used in the X Axis in the trend chart
  • Aggregated value: This column will be used in the Y axis in the trend chart

• From the dashboard you can use the logger table as data source (via GQI query). Currently trend charts supports GQI queries as data source:

Hope it helps!