Smart Split
A new Smart Split feature is available in Data Integration when importing a large volume of data that exceeds Essbase governor limits. This feature enables you to split a data load into multiple smaller loads by enabling you to define smaller data slices without having to create multiple integrations.
Applies to: Enterprise Profitability and Cost Management, Financial Consolidation and Close, FreeForm, Planning, Tax Reporting
To use the Smart Split feature, you create a base integration definition with the data to be loaded from the source application. Then you use the new Integration with Smart Split job type in the Pipeline to specify the integration name, Split dimension, Import and Export modes, Start and End periods, and Split method.
You can apply a filter for the Split Dimension as well as filters for other dimensions in the base integration. When specifying filters for the Split dimension in the base integration, you can simply provide the parent member name for the hierarchy branch to process or provide the member function ILvl0Descendants.
The filter for the Split dimension supports only ILvl0Descendant member functions. If no filters are specified, then the system processes all level 0 members for the Split dimension.
You select the method for how members in the dimension are split when the integration is executed. Split methods include: Single, Group, Custom, Custom with Functions.
In the following example, the Split Dimension has been placed on the “Entity” dimension in the source application. The Split method is “Single” and instructs the system to run a separate process for each member in the dimension.
Image shows the Integration with Smart Split job type in the Pipeline
Business Benefit: The Smart Split feature provides a scalable solution to loading a large volume of data. Use this feature when you need to split data loads due to governor limits without having to create multiple integrations and batches.
Key Resources
- In Administering Data Integration, see: