Installation: Installer And Configuration

Reviewed the Primavera Data Warehouse Installation and Configuration Guide and the Primavera Analytics Installation and Configuration Guide.

Set the JAVA_HOME and PATH definition.

Note: The config.cmd or config.sh needs JAVA_HOME set in order to launch. For example, JAVA_HOME=C:\Program Files\Java\jre1.8.0_72 (with Windows) or PATH =$PATH:/u01/jre1.8.73/bin (with UNIX or Linux).

Determine your date range for your Primavera Data Warehouse.

Note: Determine your date range before running the installer. This date range will include spread data and the start date and rolling date range of your data warehouse. See the Primavera Analytics On-Premises Administration Guide for details.

Determine a reasonable date range setting.

Note: If you set a large data range this will result in a very large amount of spread data. A large amount of spread data requires more database space, more hard drive space for temp files, and increased run times. Consider your date range carefully. Spreads outside of the date range will combine into one bucket at the beginning and end of the date range.

Determine your Project Trend Interval.

Note: Project Trend Interval will determine how to store your project history. See the Primavera Analytics On-Premises Administration Guide for details.

Determine which activity, project, and resource codes you will use before installation.

Note: Dynamic codes are available for slicing data in OBI. Code hierarchies are also supported if you select them when you install this application.

• Matching Criteria/Regular expression: The name of the code as registered in Project Management.
• Name: The display name for the code name label in OBI.
• Description: The display name for the code value label in OBI.

Determine which activity, project, and resource UDFS you will use before installation.

• Matching Criteria/Regular expression: The name of the UDF as registered in P6 and/or Primavera Unifier.
• Name: The display name for the UDF name label in OBI.
• Description: The display name for the UDF value label in OBI.

Consider the available disk space when you set the Logging level.

Note: If DEBUG is set, the logs can become large. Oracle recommends INFO for a logging level unless troubleshooting is an issue.

Set the Java Max Heap setting to at least 4 GB of memory. Allow for more memory, if available, for the Java process.

Note: The default for the Maximum Java Heap Size (MB) parameter is 1,204MB. The minimum value is 512MB. The value assigned must be a multiple of 512MB. This guide references 1 GB, but notes that 1 GB may be inadequate for many datasets and may cause failures in the ETLCalc process. Oracle recommends starting with a minimum of 4 GB of memory for the Java process.

Have the P6 EPPM and/or Primavera Unifier privuser username/password information available.

Have the Primavera Data Warehouse instance and system username/password information available.

Determine which P6 EPPM and/or Primavera Unifier users will be reporting users.

Note: Reporting users will have a database user created for them. That means you must set the report_user_flag field to ‘Y’ on the USERS table in the P6 EPPM and/or Primavera Unifier database for all those users who will become reporting users.

Define Financial Periods before running Primavera Data Warehouse processes.

Note: If you will set the Project Trend intervals based on the Financial Period, you must set the Financial Period before running Primavera Data Warehouse processes.

Set the Bulk Load files location for the ETL process to a location with a large amount of available disk space.

Note: These files can become large depending on the size of the database. You need ranges from a few gigs to many tens of gigabytes of available space.

Set Oracle database user requirements for users and passwords when creating the Oracle stageuser and staruser.