The mapping in a Kibana connection determines which model information is exported to Elasticsearch. Below you can find a few examples possible mapping scenarios for your Kibana mapping. Which scenario is suitable for your Kibana connection, depends on the information that you want to unlock and the way the model data is indexed in Elasticsearch.

Mapping with a single object type

The most straightforward way of exporting object data is using a single object type. You only use one metaobject in your mapping and add attributes and/or metrics to it and map them all with the document definition.

Data of a single object type is exported to an Elasticsearch table.

Mapping with related objects and relations using related objects' attributes and metrics

A mapping with related objects and relations can be used if you want to include relationships in your object data. The relations from the mapping will be added to the Elasticsearch table as properties. The metaobjects in the mapping form a related object chain. When mapping the metaobjects and their attributes/metrics to the document definition, the related objects are mapped to the document definition as nested documents.

Data of multiple object types is exported to an Elasticsearch table.

Mapping with related objects and relations using derived attributes

Defining related objects and relations in a mapping can also be realized by using a single metaobject and include derived attributes to link to related objects. Using derived attributes can be an alternative approach to define your Kibana mapping, but please take note that it is not possible to define metric information for the related objects. Using metric information of related objects is only possible when defining separate metaobjects for the related objects (see the previous scenario).

To decide whether you want to use derived attributes in your Kibana mapping:

  • Find out if using derived attributes will deliver the correct data to your Elasticsearch table. It means that you may need to try both approaches and see which one has the desired results in the table.
  • Determine whether you want to disclose metric information for the related metaobjects. If so, you can better use multiple metaobjects and metarelations.

The example mapping below uses the same metaobjects as the mapping in the previous scenario, but instead of separate metaobjects and relations, derived attributes () are included to refer to these metaobjects.

When adding a derived attribute, navigate from the source metaobject via the relation path to the goal metaobject (via the intermediate metaobject) and select its attribute. Following the above example, for the bottom derived attribute ( the path to follow would be:

Application component > serves > Business process > realizes > Capability > Basic profile > name

Data of multiple object types is exported to an Elasticsearch table, but in a different structure then when using multiple metaobjects.

Elasticsearch and Kibana are trademarks of Elasticsearch BV, registered in the U.S. and in other countries.