Friday Jun 14, 2013
Thursday Jun 13, 2013
By Jeffrey Mcdaniel-Oracle on Jun 13, 2013
In P6 Reporting Database and P6 Analytics versions 3.1 a metadata feature was added to allow for altering of calculations and a way to add new calculations. There is already an existing path for adding new scripts - user_scripts. User_scripts will allow you to establish new steps in the ETL process such as - adding additional tables or fields and custom calculations to those new objects. The metadata feature allows you to change calculations on existing fields in historical, burndown, or workplanning fact tables.
See documentation for more information:
Validating the metadata before running the ETL is key. If the changes you added will fail, the ETL process will revert back to a default calculation for all metadata calculations. For example, you change the metadata calculation on two fields. One of the calculations causes a failure, both will revert back to the default calculation column for what to run. Here are the steps to validate your metadata calculations:
- After making change. Run StarETL for steps 17 through 20
- Linux: staretl.sh -from 17 -to 20
- Windows: staretl.bat 17 20
- Check the log output for any errors or warnings.
A few more items to be aware of:
- Calculations are per data warehouse/not per data source
- Drop STARUSER and you will lose all custom calculations
- Metadata calculations are stored in the ETL_CALCULATIONS table
Provide new information on Primavera Analytics and Data Warehouse
- Upgrading Older Analytics\Data Warehouse Versions
- Parallel Loading
- Spreads in Primavera Analytics
- Having Trouble Publishing Projects?
- When Configuring Multiple Data Sources...
- I Created an ETL Schedule Through the Web Configuration, How Do I turn the Schedule Off?
- Summary Data and the Summarizer, What is Needed for P6 Analytics?
- Capturing Historical Financial Periods for Analytics
- How Could I Get a Project to Publish After a Certain Amount of Activity Code Changes?
- Why Aren't My Code Changes Being Reflected in Analytics?