Schemas are data models written in the Pantheon Schema Language. Schemas define logical views of data that allow you to integrate your physical data without the need to duplicate or copy it. Schema definitions likewise enable you to combine multiple data sources in a single schema definition.
Schema definitions consist of:
Physical mappings that define schema tables and schema columns; and
Logical models that define dimensions, measures, nested schemas, and conformance definitions between dimensions.
The physical mapping of a schema defines schema tables and schema columns. It represents an abstracted view (or mapping) of physical data from an underlying data source. The mapping can provide access to one or more data sources, and it provides uniform access to all referenced data sources via standardized SQL queries.
Schema tables are abstracted views of physical tables. A schema table is derived from either an entire physical table or one or more physical columns. Schema tables contain schema columns.
Schema columns are constituents of schema tables. They are derived from physical columns.
Tables from underlying physical data sources.
Columns from physical tables in underlying physical data sources.
Logical models within schemas define dimensions and measures derived from the physical mapping with schema tables and schema columns. Dimensions and measures aggregate, compute, and contextualize data, which allows you to execute multi-dimensional queries.
Measures are aggregates of schema columns that quantify your data and return numerical values.
Dimensions define a namespace for their constituent elements, or dimension attributes. Dimensions are derived from schema tables and represent an additional way to categorize your data. They are qualitative and provide context necessary to understand the meaning of measures.
Elements of dimensions, or dimension values, added to queries to break down measures.
The physical mapping and logical model together form the abstraction layer. The abstraction layer provides a logical view on physical data. This enables Contiamo Data Hub to transform and integrate data without having to copy, duplicate, or move it. All transformations and integrations happen on the logical level, with any changes to the data itself happening directly at the data source. The abstraction layer allows Contiamo Data Hub to create a logical view on complete data sources and creates a centralized access and management location for all of your data.
Writing a schema creates a logical view of physical data through an abstraction layer. Using Contiamo Data Hub, you can integrate these logical views and create uniform, logical data sources that exist on top of your physical data. Logical data integration creates uniform views of your physical data without creating duplicated stores and allows you to centralize and individualize access rights and permissions.