BI Blend uses in-memory processing and does not write-records to stage tables. All results are output to the designated target database. This engine is used to rationalize the transactional data by leveraging the OneStream Metadata Engine. During processing the BI Blend Cache will utilize the Cube dimensions to transform the records into a unified reporting format. The Cube Hierarchies are used to derive aggregation points to be stored for reporting.
The BI Blend Stage Cache will also accept properties on Entities and Accounts to perform direct method translation to the application reporting currency. The Engine ultimately generates a finalized table with a Column Store Index, creating a structure for OLAP reporting. The table can be accessed with the standard OneStream BI Dashboard and Pivot Grid reporting tools for analytic reporting as well as Relational Blending into Cube Views and Dashboards.
The BI Blend Stage Cache Engine supports:
-
Hierarchy Aggregation without complex calculations
-
Simple currency translation using the Direct Method only. Any destination currency can be defined, but only one per BI Blend process.
-
Limited use of member properties
-
Does not use dimension relationship properties such as Aggregation Weight, Percent Consolidate, or Percent Ownership
-
Limited parent level calculations using derivative rules
-
Basic time math using helper rules
-
Supports simultaneous multi-period data loads by record for up to 12 periods using Attribute Value dimension Members
The Stage Engine integrates transactional data. Source data is processed through the standard Stage Engine, where the data uses the common OneStream tools to parse and transform records. BI Blend source data contains additional members that are defined in the integration as Attribute members. Stage Derivative Rules enhance the data with groupings available in reporting.
