Data Flow

When transferring statistical data into the Business Warehouse, a central monitoring system (CEN) with a release as of SAP NetWeaver 7.0 SP3 acts as a data source. The statistics data itself originally comes from satellite systems that are defined as monitored systems in the monitoring infrastructure (see Registering SAP NetWeaver Components and Hosts in CEN).

The connection to the Business Warehouse is created by the PI BASIS Plug-In as of 2005.1 7.00. This plug-in contains the extraction function modules.

Finally, the BW system requires release SAP NetWeaver 7.0 SP 9 with BI Content as of 7.0.3 Add-On SP 2. This contains a range of data containers for the statistics data. Queries for creating reports are defined in relation to these data containers.

This graphic is explained in the accompanying text.

Process of the Data Flow

The data flow is processed in the following steps:

The single records are read from the monitored systems using the extractor for statistics data. This data is transferred using the RFC destinations that are defined in the monitoring infrastructure of the CEN.
The read data is initially stored in Persistent Staging Areas (PSAs); these database tables represent the inbound store in the BW. There are separate PSAs for statistics main records and for each statistics subrecord type.
The statistics data is transferred from these PSAs into the following objects:
- The data is transferred in the format for the Single Records to Data Store Objects; there are separate DSOs for statistics main records and for each statistics subrecord type.
- The data is transferred in the format for aggregates to various InfoCubes. There is a separate InfoCube for each aggregate (which are based on the Analysis Views of the Workload Monitor).
  
  Note
  
  The Configuration Data for Web AS Statistics Data and Configuration Data for DSRs determines the retention period of the individual records and aggregates in the BW.
  
  End of the note.
A single statistics record always consists of a main record and any number of subrecords that contain additional, specific information about the business action of the main record. This is reproduced in the BW, in that main records and the corresponding subrecords are combined into InfoSets.
If queries request data for single statistics records or aggregates, the request is met by the following objects:
- In the case of aggregates, the corresponding InfoCube is read.
- In the case of single records, a MultiProvider satisfies the request. If the desired data already exists in the BW, the request is forwarded to the corresponding InfoSet. If the data does not yet exist in the BW, the extractor for statistics data collects it directly from the source system and provides the data in the BI using a RemoteCube.

Statistical Data start page