What is a data model?

When storing data, the main thing is to define the layout of the data in such a way that it can still be accessed years later. Therefore, for a future-proof data management system, it is essential to define the entities with their attributes, constraints, and relationships to each other to ensure future data understanding. 

Definition of entities and relationships

Defining entities and relationships is not new. Entity relationship models have been around more or less since the beginning of the relational database era. And the key parts are: 

• Defining the entities that exist in your data management system
• Defining the attributes of those entities with their corresponding data types and constraints
• Defining the relationships between those entities

Entities, including their attributes and relationships, are sometimes also referred to as metadata. 

From entity relationship models to a data model

Once entities, attributes and relationships are defined, the next important question is: what do these entities and attributes mean? For example, if you need to identify the entities that are used to define a unit catalog and which of the associated attributes are factor and offset in case you want to perform a unit conversion on your measurement data. 

In this case, semantics comes into play to define the meaning and interpretation of the data. Furthermore, by adding a data context and catalogs for recurring content, you have already defined a data ontology – and all of this together is what we call a data model. 

Use the ASAM ODS industry standard for defining data models

The aforementioned ontology can be expressed by introducing what we call a base data model, from which you can derive your own application data model. 

Defining your application data model in this way allows your entities to be derived from predefined base entities and to inherit predefined base properties, thus solving the problem from above of identifying the entities and attributes for defining a unit catalog data model.

An industry standard that provides a basic data model and rules for deriving an application data model is ASAM ODS, which can be seen as an ontology for test and measurement data management.