The imported data is also saved persistently in the PPM database. The structure of the analysis realm tables in which the data is saved is defined in the analysis realm configuration.
A change to the transformation factors (<client>_transformationfactors.xml) is therefore transferred when the analysis server is reinitialized.
To avoid having to initialize data after every server start, the analysis realms of the analysis server are saved in recovery files and loaded automatically.
Data import with duplicate keys
If a table row is imported and has the same key as an existing row, the existing row is replaced by the new row.
If the import file contains multiple rows with ab identical key, only the last row is imported and saved in the database.
Data import with missing criteria or criteria that cannot be calculated
If a criterion for a row cannot be calculated, an error message is output and the row is not imported. The same applies to key and non-key criteria.
If a column is missing in the import file or if a column is empty, the criterion based on this column is not imported. In case the criterion is used as a key and it is not a text dimension, the entire row is skipped because the missing key must be an error.
Reinitialization behavior differs from data import behavior: If a measure or dimension cannot be calculated during reinitialization, an error message is output, but the relevant row is transferred to the analysis server. A text dimension is assigned the default value. All other criteria are handled as "not specified". This does not apply to key criteria.