Data Models

DataMosaix has two basic types of data models. There is a simple Asset Hierarchy and Flexible Data Models (or just Data Models).

Asset Hierarchies is a simple parent/child relationship where each asset can have an unspecified number of properties and have child assets. There are no specific types. Each asset can have associated events such as time series, documents, and more associated with them. This is the older data modeling method in DataMosaix. It has the advantage of being conceptually simple, but there are some limitations with modeling objects other than assets.

The newer Flexible Data Modeling has been added by Cognite over the last year. They are planning to migrate everything to it eventually. It is a much more flexible modeling system based on GraphQL and has the ability to define types (such as asset, equipment, pump, peristaltic pump, and more) with specific properties. The major advantage over the Asset Hierarchy is you can define types for things that are not assets with different relationships. For example, you can define a work order type with a relationship to spare parts, the technician, and hourly rate.

Some of the industry applications use the new data models (BPA and AI for Mining) and some use the Asset Hierarchy (Energy Manager). The industry solutions teams plan to have a new base data model that is shared and extended by each app in the future.