Setting the data warehouse run parameters

Set the data warehouse run parameters to initialize the data warehouse run control table and establish the appropriate run parameter settings for executing the data warehouse ETL process at regularly scheduled intervals. You can also modify run parameters as necessary and execute an ETL process manually. Set parameters to initiate a full or partial (incremental) data load and/or purge data.

When running the ETL process for the first time, you must run a full load with no errors. After running a clean full load, you do not need to run a full load again, because the data warehouse retains all of the data from the full load. Subsequent future loads can be run automatically based on your data warehouse setup, e.g., on a nightly basis. Future loads will be incremental and will only add to the existing data by loading the data since the last load. Therefore, after completing a full load, the system automatically resets the Load from Date to the current date (date of the most recent load) and the Load Type to Incremental so that you do not have to reset those parameters for future data loads.

Note: Although an incremental load only loads data since the most recent load, an incremental load also verifies time stamps for records so that it will load any changes that have been made to an existing historical record or for transaction data that may be post-dated, etc.

Although it is not necessary, you can run a full load if desired. If you select Purge Old Data, running a full load purges all existing data in the data warehouse before loading any data; therefore, you can run a full load to purge the existing data from the data warehouse and then reload all of the data based on a new Load from Date, etc.

Note: The data warehouse does not maintain historical values for parts and stores. Instead, the data warehouse takes daily snapshots of the value of materials and inventory in each snapshot. Therefore, purging the existing data from the data warehouse will also purge non-historical value information.

See Understanding historical inventory valuation for the data warehouse.

After running a full ETL load, you must check the data warehouse run log for any errors.

See Viewing data warehouse run log errors.

You must run a full load without any errors prior to switching to an incremental load. If any errors occur during the full load, you must correct the errors and run a full load until no errors occur.