Live Data Connect for BOBJ - Where is the data?

[laura_vega]

LDC for Universes and Web Intelligence documents Overview

After describing Live Data Access on top of Business Objects Universes and Web Intelligence documents in previous blogs, let's focus today on the technical aspect of this access, and more specifically on the data itself.

As a reminder, LDC for BOBJ allows you to connect from SAP Analytics Cloud to your relational Universes or to your documents data models, and to create stories and applications on top of this valuable data.

Note this post only describes LDC access for Live connectivity mode, not acquisition mode.

 

Architecture

This schema represents LDC architecture, and how LDC module (or add on) s connected to your BOE system and to SAC.

Let's describe each access type and which components are used in each workflow.

 

LDC for Universes Workflow

1. Create Model
   LDC connects to the CMS to retrieve all available universes you are allowed to use. In the Query Builder, the Universe layout is displayed, you select the objects to add in the model.
   When clicking OK, no data is retrieved yet. SQL is generated by LDC and the query definition is serialized, stored along with the model.


2. Create a story
   When creating a new story using this LDC/Universe model, the first action you will perform requiring data display (create viz, data viewer), will send the SQL to LDC which will delegate it to BOE connection server (CS). The query is executed on the DB and the data stored back in a transient WebI microcube (Cube) in LDC module.
   After this first access to the DB, all further data needed in the SAC story will be computed by the LDC cube whenever aggregations or calculations are needed from the story contents.
   Only the data needed in your visualizations in the story is displayed in your browser. There is no data movement to the cloud.


3. Open an existing story
   When opening an existing story based on a LDC/Universe model, a new SQL refresh for the whole model is executed as no data is stored in the saved story. All existing visualizations are refreshed with the transient microcube recreated in LDC module.
   Then any action will be similar to the ones described in step 2.

This workflow will explain you why some universe features are supported by LDC access, and some others not.

•  For instance, all data security settings set at Universe level in the Data Security Profile are supported by LDC. This is because the query generation is performed by the LDC module which uses the user credentials and thus the BOE user rights.


• On the contrary, delegated measures (measures which aggregations which are not performed by the microcube but sent to the DB) are not supported, as there is no DB access every time you perform an aggregation in the story.

 

LDC for WebI documents Workflow

The workflow for LDC accessing WebI data models is very similar to the Universe one, except there is no access at all to the underlying DB. All the data comes from the WebI document selected in the SAC model.

Let's go through all the steps anyway:

1. Create Model
   LDC connects to the CMS to retrieve all available WebI documents you are allowed to use. In the Query Builder, the document data model layout is displayed. This data model contains all the document data providers information such as objects, variables, and synchronized dimensions. You select the objects to add in the model.
   When clicking OK, no data is retrieved yet. The query definition is serialized, stored along with the model.


2. Create a story
   When creating a new story using this LDC/WebI doc model, the first action you will perform requiring data display (create viz, data viewer), will load the webi document into the LDC add on. The model query definition will retrieve the corresponding data from the document. There is no access to the DB. The data is stored in the transient local cube in LDC add on.
   After this first access to the document, all further data needed in the SAC story will be computed by the cube whenever aggregations or calculations are needed from the story contents.
   Only the data needed in your visualizations in the story is displayed in your browser. There is no data movement to the cloud.


3. Open an existing story
   When opening an existing story based on a LDC/WebI doc model, all needed data to be displayed in the story is  first fetched from the WebI document loaded in LDC module into the cube.
   All existing visualizations are refreshed with the transient microcube recreated in LDC module.
   Then any action will be similar to the ones described in step 2.

Here again, you will understand why with this WebI document access, as there is no access to the underlying DB, following WebI documents features are not supported:

• Refresh On Open


• Query Stripping


• Delegated Measures

 

As a conclusion, LDC for Universe and LDC for WebI documents are very similar.
Just remember that for Universe access, the data source is the DB, while for WebI doc access, the data source is the WebI document itself.