Siebel Performance Management Guide > Performance Scorecard Reference >
About the Analytics Data Loader Process
This topic describes the background processes that occur when the Analytics data loader functionality is employed.
NOTE: This topic is not meant to be a step by step procedure; rather it is an overview of the basic functionality.
Request Selection in KPI Administration (Siebel Analytics Request Column)
When a user chooses a report, a pick applet appears. This applet is derived from a virtual business component (VBC).
The VBC calls a business service, which connects to the Analytics server and then passes certain parameters such as login, password, root catalog path, and type (Analytics data loader).
The user logged into the Analytics server can then browse the catalog from the root path provided for that user (all responsibilities and visibility rights for that user are enforced at the Analytics server for the user.)
NOTE: The root path for Web catalog administration is configurable using Siebel Tools. For more information, see Using Siebel Tools.
When the administrator clicks Refresh KPI, a business service waits on the Invoke Method from BusComp, and a business service performs logic on the KPI to see if it should be refreshed.
If the KPI needs to be refreshed, the business service passes the KPI parameters to the Marketing Analytics Data Loader business service (MDL).
The MDL business service passes certain parameters to the Siebel Analytics Server, including the administrator ID and password from the Analytics server configuration, the KPI ID, catalog path, and server ID.
The following MDL business service processes occur:
- Impersonate. The business service logs the administrator ID into the Analytic Server.
- Write-to-List file. The Analytics server generates a list file from the specified catalog path, and then inserts the KPI_Id into this list file.
- Because this is an asynchronous process, the job_id is sent back to the Siebel application as the write-to-list file command is occurring.
- The Siebel application polls the Analytics server to determine when this job_id task is done.
- Each time the Siebel application receives a new file from the Write-to-List file, the command Process file is called, which then calls the Import KPI workflow.
Many instances of this workflow can be running at the same time. The file is processed in the following ways, using Siebel EAI functionality:
- The process turns the Analytics-generated CSV file into a property set.
- The property set is upserted into the integration object.
The integration object is called ERM KPI Definition Employee Value EAI for employee-specific KPIs and ERM KPI Definition Admin EAI for global KPIs.
The difference between these integration objects is that the global KPI integration object receives a file from Siebel Analytics that has two columns and one row (KPI Id and KPI Value.) The employee-specific KPI integration objects receive a file from Siebel Analytics that has three columns and many rows (KPI Id, Emp Id, and KPI Value), and the number (#) of rows is the number (#) of employees that have specific values for that KPI.
The CSV files that Siebel Analytic exports are kept in the Marketing file system. The data is dropped from the integration object in the appropriate business component.
- The CSV file holds the KPI_Id, Integration object name, and Integration component name because they are part of the Siebel Analytics request.
- The integration objects insert the data into the correct business components.