Restrictions for the InfoCube with Data Persistence in the BWA

Selective Deletion/Data Archiving

Selective deletion is not available in the BWA for InfoCubes with data persistency. Data archiving is also not available because selective deletion is also required for data archiving.

Delete by request:

If you want to frequently delete by request, you should select a standard InfoCube instead of the InfoCube with data persistency in BWA. The reason for this is that selective deletion only allows data to be logically deleted from the BWA. Only the request information is only removed from the package dimension. The data is kept in the fact table. Frequently using selective deletion results in a large quantity of unused data in the fact table.

Compression:

Request compression is not available in the BWA. There are three reasons for compression. One of these reasons is not relevant for the BWA:

• Database space and query performance:

  One reason for compression is to save space in the database and improve query performance. This reason is not relevant for the BWA, as the BWA processes and calculates the data in a very compressed state anyway.

• Calculation of the marker for non-cumulative key figures:

  The marker for key figures should be regularly updated by frequent data compression, to improve query performance. You should only use non-cumulative key figures in standard InfoCubes because compression is not available in the BWA.

• Compression with zero elimination:

  Compression with zero elimination is not possible for the InfoCube with data persistency in the BWA. In such cases, you have to use a standard InfoCube.

Reconstruction of BWA index:

It is not possible to reconstruct the BWA index of the InfoCube with data persistency in the BWA from the fact table. If you need to reconstruct the index, you have to reload the data using the DTP in the Full extraction mode. Therefore we recommend that you load the data from the InfoCube into a DataStore object. You can then reload the data into the InfoCube and thereby index the data again.

Parallel loading using DTPs:

Due to technical reasons, you cannot parallel load using multiple DTPs. Only parallelization within a DTP is possible.

InfoCube Administration:

The InfoCube administration contains fewer functions, because the functions not required for the InfoCube with data persistency in the BWA are not available.

Backup and Restore:

Backup and restore functions are not available.