An Introduction to the Data Warehouse

  •    The warehouse design is based on Ralf Kimball’s data model
    • The Kimball model includes both dimensions and facts
    • The model revolves around numbers (facts)
    • The dimensions describe the facts
    • A dimensional model is optimized for reporting (queries)
  • The warehouse is a “snapshot in time” of Banner data and other non-Banner data sources
    • example:  4th week data
  • Cognos is the reporting tool that is used to pull the data from the warehouse
    • Converted as PDF and printed
    • Saved as Excel file
    • e-mailed to your inbox

  • Facts (also referred to as cubes):
    • contains the data corresponding to a particular business process, action or event; typically numeric (links two or more dimensions together)
    • example of a fact: registration(credit hours, grades, quality points)
  • Dimensions:
    • contains details about each instance of an fact  (descriptors)
    • examples of dimensions: academic year, academic period, entity(last name, first name), section(course title, course number, crn, section term code)

 

diagram


  • The Data Warehouse was placed into production on June 2, 2014.
  • The warehouse includes 108,702,145 rows of data from Banner Finance, Student and Human Resources and Financial Aid
  • Reporting includes pre-built and ad-hoc queries.
    • Pre built reports will include parameters that you will respond to at run time
    • Ad-hoc queries will give you the ability to have full control of the report
  • The data is refreshed on a nightly basis.
  • # of Dimensions ………………………….72
  • # of Cubes ………………………………….56
  • # of data elements (columns) …..2,666

Demonstration:

For access to the data warehouse, contact Faye Whitenack at whitenaf@wou.edu.

For Cognos Warehouse training, contact Max Chartier at chartierm@wou.edu

 

Data Warehouse

The new dimensional data warehouse was launched June 2, 2014.  The WOU model uses the Kimball approach to modeling.  The current warehouse contains data from Finance, Student and Human Resources.  Both canned reports and ad-hoc query capability are available.  The data is refreshed on a nightly basis.

The model was developed using a product called Oracle ODI.  ODI is the same tool that is used to perform the nightly data refresh.  Historical data is preserved during the refresh process.  The modeled data is then transformed into a business friendly format utilizing Cognos Framework Manager.  Cognos Connection is the web-based end-user reporting tool.

The warehouse team includes:

  • Richard Kavanagh — Kimball modeling expert
  • Shea Hawes — SIS expert
  • Nick Miller — FIS/HR expert
  • Max Chartier — Reports guru
  • Jared Petersen — Reports, web docs, etc.  (student employee, honors student)
  • Michael Ellis — Project development

For access to the data warehouse contact datasecurity@wou.edu

Data warehouse resources can be found here.

Available warehouse reports can be found here.

The warehouse can be found here.