How to Create an OLAP Cube Using a Management Pack


Updated: May 13, 2016

Applies To: System Center 2012 SP1 - Service Manager, System Center 2012 R2 Service Manager, System Center 2012 - Service Manager

A System Center online analytical processing (OLAP) cube is defined as a collection of the following management pack elements:

  • Measure groups, which points to the facts that are included

  • Substitutions to be performed on the fact

  • MDX resources

  • Drill-Through Actions

  • KPIs

  • Named Calculations

  • Custom Measures

  • Many to Many Relationships

All elements other than MeasureGroups are optional.

When you define an OLAP cube using the schema above and you import the management pack, the System Center 2012 – Service Manager data warehouse deploys the cube using SQL Server Analysis Management Objects (AMO), and it creates the required infrastructure to maintain it. For example, you can create a simple OLAP cube based on ComputerHostsOperatingSystemFact. The following illustration is the dimensional view of the fact in the warehouse.

Diagram of dimensional view

To create an OLAP cube using a management pack

  1. Copy the following management pack source code and save it:

    <SystemCenterCube ID="ComputerCube">  
    <MeasureGroup ID="ComputerHostsOperatingSystem" Fact="DWBase!ComputerHostsOperatingSystemFact" />  
  2. Import the management pack, and then run the MPsync job. The OLAP cube will appear in the Service Manager console in an unprocessed state.

  3. To view OLAP cube, open the Service Manager console, navigate to Data Warehouse and Cubes, and then click ComputerCube.

  4. A data warehouse process job is created for the OLAP cube with a default 24-hour job schedule. Therefore, process the cube using the Service Manager console or using the cmdlet Start-SCDWJob -JobName Process.ComputerCube.

  5. Open the cube in Excel using the link from the Task pane, and look at the cube structure that was created.

  6. Notice that the following measure groups are created for the OLAP cube:

    • A measure group corresponding to the fact ComputerHostsOperatingSystemFact with a Count measure

    • A measure group corresponding to the dimensions that it points to

    • Computerdim and OperatingsystemDim with the count measure

  7. Notice that the following cube dimensions are created:

    • The outrigger dimensions corresponding to the fact are added as cube dimensions so that you can slice the facts on those dimensions. These dimensions include Priority and Status.

    • DateDim is added to the OLAP cube because it is relevant to any fact.

    • EntityStatus and RelationshipStatus cube dimensions are defined for all cubes to indicate whether the entity or relationship is deleted.

See Also

Understanding OLAP Cubes