This page provides information on how to configure IFS Analysis Models with respect to executing user, security, basic data, Access Views and data tranasfer.
To make the transfer of information from IFS Applications to the Analysis Models specific Data Warehouse in SQL Server successful, it is important to consider the available configuration options.
The information can be used both before performing the first load of the Data Warehouse in SQL Server as well as to supply general configuration guidelines for subsequent loads.
Some general issues as e.g. translations are also handled.
Transferring information from IFS Applications to SQL Server requires a user in IFS Applications with necessary permissions and access rights. This version of IFS Analysis Models strongly suggests the <IFSINFO> user as the executing user.
Note: The <IFSINFO> user will have all necessary views available in his own schema in the database and the user will also have full data access to all views with security filters. The <IFSINFO> user also needs to have access to some packages in the <appowner> schema and this is handled by creating private synonyms for the <IFSINFO> user.
The executing user also controls the language that will be used. The following is important to configure before extracting any data:
This section deals with data object access and data access related to IFS Applications and also provides some information about security handling in SQL Server. Please refer to:
More about how translations are handled can be found here >>
IFS Analysis Models requires that basic data has been set up properly in IFS Applications.
Extracting data from IFS Applications to the Data Warehouse in SQL Server is done by accessing Access Views. These views represent the read interface in IFS Applications, acting as a view layer on top of the Information Source related views.
Note: Access Views are not available by default. It is necessary to generate these views via the IFS EE client.
Learn more about Access Views:
The ETL process in SQL Server is dependent on so called Lookup Data, stored in a set of lookup tables in the SQL Server database. It must be made sure that these tables are correctly defined. Please refer to Basic Data in IFS Applications. Learn more about lookup tables >>
The loading of the stage area in the Data Warehouse can be handled in different ways. More information about the ETL process can be found here.
IFS Analysis Models supplies possibilities to configure the load volumes. This is especially important for the initial load of fact related data but also for the subsequent loads in order to be able to reduce the load time. Learn more >>
If the ETL process related to IFS Analysis Models is scheduled, it can be of great interest to know when the job has finalized and what the status was. One way of handling this is to configure a an E-mail to be sent when the SQL Server Agent Job finalizes.
Read more here.
To learn more about Excel Dashboards, please refer to >>
Customization of Analysis Models is not supported by any modeling tools. If changes are made in existing Fact or Dimension access views (online/data mart) it will be necessary to generate new Access Views and to update tables, procedures, SSIS packages and SSAS models manually.
If the customer versions of the staging SSIS packages contain calls to PL/SQL packages in the IFS Applications database, packages that are not used by the core versions of the SSIS packages, then it is necessary to make sure that the <IFSINFO> user has access to these packages. There are two things to consider here:
INSERT INTO BIBASE_REF_PKGS_TAB (PACKAGE_NAME, INSTALL_DT, CORE_DEFINED) VALUES (<pkg_name>, SYSDATE, 0)
BEGIN BIAXSV_INSTALLATION_API.BI_POST_INSTALLATION_OBJECT; END;