When you created the Primavera Data Warehouse schema and enabled partitioning, the Configuration Utility created table objects using a list partition based on the number of data source IDs identified on the Partitioning ETL Settings page. If you require additional data sources in the Primavera Data Warehouse schema beyond the total amount of partitioning data source IDs that you originally allocated, complete the following steps:
Note: You can confirm the amount of list partition counts in your Primavera Data Warehouse schema by connecting to the database instance using the Data Warehouse schema owner (by default, STARUSER) and running the following statement:
select unique partition_count from user_part_tables where partitioning_type = 'LIST';
- Back up your existing Primavera Data Warehouse schema.
- Do the following to use the Primavera Data Warehouse Configuration Utility to increase the number of list partitions:
- Go to <PDW Install Folder>\star\config.
- Run configStar.cmd (with Windows) or configStar.sh (with UNIX or Linux).
- In the Configuration Utility, go to the ETL Settings page.
- In the Datasource ID field, enter the number of list partitions you need.
Note: You can set up to 99 list partitions. The set value can exceed the amount actual data sources you need.
- Do the following to recreate your Primavera Data Warehouse schema to include the increased number of list partitions and repopulate your original data source data:
- Run setup.bat "-r" "-u" (with Windows) or setup.sh -r -u (with UNIX or Linux) from the first data source folder to recreate the Primavera Data Warehouse tables with the new number of partitions. Note that, at this point, the history tables have been moved to backup tables and your schema has been recreated with data from your first data source only.
- Go to <PDW Install Folder>\star\etl_homes\staretl<id> where <id> is the identifier for your next data source.
- Run staretl.bat (with Windows) or staretl.sh (with UNIX or Linux) to repopulate Primavera Data Warehouse.
- Repeat steps b and c for each data source.
- After all of the original data source have been recreated, locate the W_PROJECT_D table and verify that there is a datasource_id =<data source ID> for each data source.