Data Warehouse Project Description

Definition:

The Data Warehouse Project consists of two primary components: building a data warehouse and providing an end-user query/reporting front-end to the data warehouse. The data warehouse (DW) is a separate database built on a separate server. This keeps the resource intensive report generation from interfering with the important on-line processing such as registration, payroll, or cashiering. The DW is built by copying the data from the production Banner database, WWIS on UIS (previously known as Hydra), reformatting the data and loading it into the data warehouse server. The front-end query tool, Hummingbird BI-Query, is configured to present a graphical representation of the data in the data warehouse in an easy to understand manner.



Purpose:

The purpose of this project is to present central administrative data in an understandable format, allowing flexible presentation, and in a way that is easy for users to access



Objectives:

  • Format the data such that it is easier for users to understand and to access.
  • Deliver a set of pre-written reports.
  • Provide a reporting tool that allows users to write their own adhoc reports.
  • Easy to use
  • Flexible
  • Powerful (Support crosstabs, prompts, master-detail, charts, etc.)
  • Provide the users with and easy method to modify, save, & run their own reports and the pre-written reports.
  • Deliver the above while reducing the load on the strategic production machine, so that student registration, cashiering, and payroll continue to perform well, even as the demand for more reports grows.
  • Provide training for users in the use of the front-end and in understanding the data structure.


Specific Goals:

  • Build a data warehouse of the most commonly used data on a machine other than the production machine.
  • Develop routines to update this data regularly from the production database.
  • Reorganize the data tables into few tables with an easier to understand format
  • Rename tables and data fields to be more easily understood
  • Pre-define table relationships so users do not need to understand or define them – just click on fields when building a report.
  • Build a user-friendly graphical front-end to the data warehouse.
  • Install and configure Hummingbird’s BI-Query end-user query tool
  • Design Bi-Query models specifically to make it easy for users to access data and write their own reports.
  • Develop a set of commonly used pre-written reports.
  • Provide an easy method for users to develop their own reports and have easy access to both the common reports and their locally developed reports
  • Human Resources delivers user training in “Writing BI-Query reports against the Data Warehouse”.
  • Institutional Research provides supplemental training for users in understanding the data.
  • Implement BI-Web so users can run reports from a Web browser.
  • Evaluate an advanced analytical tool for use by limited number of users (may be the same as above).
  • Continue to add more data tables and reports over the next couple of years.


Status:

  • Student Information Data Warehouse is in production with BI-Query Model built. The model contains hundreds of pre-written Bi-Query reports for users to run.
  • Human Resources Data Warehouse is in production with BI-Query Model built with user reports available.
  • Woodring Student Information Data Warehouse is in production with BI-Query Model built and specific Woodring College of Education user reports available.
  • Alumni Data Warehouses are in production with BI-Query Model built. Due to differences in analysis of data this model received from another Banner institution needs to be re-designed.
  • Data Warehouses have been started in the area of Admissions, Housing, and Financial Aid.
  • Aggregate Student Information System statistics are being constructed in another Data Warehouse.
  • A large number of Reports have been written and placed on a menu in the BI-Query Model and can be run by most Banner users.
  • Access is granted by the Banner Information system custodian of data, the Registrar, HR Office, Foundation, etc. on a need-to-know basis.
  • Refer to the Data Warehouse FAQ (frequently asked questions) Web page here for additional information on how to get access.
  • HR Office's Training division delivers BI-Query classes for all employees.
  • Peak Student data is capture every term and stored in the Data Warehouse for trend analysis
Page Updated 05.15.2013