Load data

You can batch load actual and target data into the plan using the Data Loader utility. The data must be either in a plain text (*.txt) or comma-separated format (*.csv) file. You can load new measures for a plan object and update existing measures where data source is defined as Manual. Measures in the database where the data source is SQL or CPM will not be overwritten.

To load data, you cannot have any data in Strategy Management's writeback table. You must run the Merge Writeback Table utility first before attempting to load data using either the Data Loader or the DTS package. If you try loading data with either method when data needs to be written back, a warning will display telling you to run the Merge Writeback Table utility first.

• You can load actual and target data only for an existing plan object.
• The Measures Source data source can be either Manual or None. It cannot be SQL or CPM. After successfully loading, None is changed to manual.
• If the Consolidation Method is Sum, the Data Loader will not load non-leaf data.
• The data will not be shown in the plan until the plan object has the corresponding unit association. You must associate the plan object and unit.
• The Data Loader will not set the measure’s requirements, such as the measure type; the direction; whether measured values accumulate over time or not; and the consolidation method.
• If you run the DTS package from the Microsoft SQL Server computer, the computer must have access to the data source file and log file. For example, say a planner creates and saves a package to a Microsoft SQL Server computer named PD-Server but the data source file and the log file are located on the planner’s computer. The directories on the local computer that contains the data source and log files must be accessible from PD-Server if you run the package from there.
• The Error.log file is created in the Program Files (x86)\Infor\CPM\Strategy Management directory and is appended afterwards. If the log gets too big, you must delete some of its contents.
• If you are working in a mixed Microsoft SQL Server version environment, you must add the Process cube task manually, outside of the package.

When constructing your data file, keep these in mind.

• We strongly recommend that you organize your data file by ID instead of by name. This avoids problems with duplicate member names since members cannot be shared. If you have a duplicate member name, that member and all of the data that follows are not loaded.
• The IDs or names in your data file must exist in the dimension tables. Otherwise the record is not be loaded.
• Each row represents a record. If more than one record is in a row, the file is invalid.
• Blank rows are not allowed. If there is a blank row in the data file, the loader only loads data up to the row before the blank row. The data rows after the blank row are not loaded.