Database Reference
In-Depth Information
4.1 The BW cleanup 12 step-program
4.1.1 Clean up the PSA
Data that has already been loaded to DSOs can be removed from the persistent staging area
(PSA). Since the PSA is no longer needed in an SAP BW on HANA system, there is no need
to migrate data from the PSA to the new system.
4.1.2
Delete aggregates
Since the HANA calculation engine can calculate aggregates on the fly, summary tables
are no longer necessary. Delete the aggregation tables which were required to provide en-
hanced report performance in BW systems which used the older layered scalable architec-
ture (LSA). Aggregate tables are no longer necessary in LSA++.
4.1.3
Compress E and F tables
The E and F tables are the fact tables for your InfoCubes. The F table stores uncompressed
data, while the E table stores compressed data. As part of the compression process, the re-
quest IDs are not stored in the E fact table. As part of the BW cleanup before a DMO migra-
tion, the E and F fact tables should be compressed in all InfoCubes to reduce the size of the
InfoCubes and the overall migration.
4.1.4
Remove statistical cube data
Your BW system saves statistics data so that it can be used for performance tuning. Stat-
istical cube data is stored in tables with technical names beginning with 0CTC_xxx. Since
the statistics data from your old relational database will be inaccurate when the migration to
SAP HANA is completed, to reduce the size of the migration, you can purge the statistical
cube data as part of the BW cleanup process before beginning the migration.
