About a Semantic Model's Architecture
The semantic model contains three layers of metadata that build on each other and prepare the data source's data for users to query and analyze.
Physical Layer
This is the first layer of the semantic model.
The physical layer defines the objects and relationships that the Oracle Analytics query engine needs to write native queries against each physical data source. You create this layer by importing tables from your data sources into physical databases and then creating alias tables to obfuscate the actual database tables from end users and to control access and updates to the actual data.
Separating the logical behavior of the application from the physical model provides the ability to federate multiple physical sources to the same logical object, enabling aggregate navigation and partitioning, as well as dimension conformance and isolation from changes in the physical sources.
The physical layer can contain: subject areas, folders, localization configuration information, role-based permissions, row-level security, variables, governance rules, and mappings.
See What is the Physical Layer?
Logical Layer
This is the second layer of the semantic model.
The logical layer defines the logical model of the data and specifies the mapping between the logical model and the physical schemas. This layer determines the analytic behavior seen by users, and defines the superset of objects and relationships available to users. The logical layer hides the complexity of the source data models.
Each column in the logical layer maps to one or more columns in the physical layer. At runtime, the Oracle Analytics query engine evaluates Logical SQL requests against the logical layer, and then uses the mappings to determine the best set of physical tables and files for generating the necessary physical queries. The mappings often contain calculations and transformations, and might combine multiple physical tables.
The logical layer can contain: business models, business entities, levels, measures, filters, aggregates, table sources, hierarchies, level-based measures, calculations, aggregates by dimension, and fragmented or federated logic.
See What is the Logical Layer?
Presentation Layer
This is the third layer of the semantic model.
The presentation layer provides a way to present customized, secure, role-based views of a logical layer to users. It adds a level of abstraction over the logical layer and provides the view of the data seen by users building requests in Oracle Analytics and other clients. The presentation layer allows users to easily query data without having to understand the underlying data source.
You can create multiple subject areas in the presentation layer that map to a single logical layer, effectively breaking up the logical layer into manageable pieces.
The presentation layer can contain: tables, data sources, aliases, joins, connections, fragmented sources, and federated sources.