Deans' Council

Presentations by ITS

Tuesday, 02 September, 1998

2:00 PM, AD316

 

 

What is Data Warehousing? – David Heise

1 What is it?

2 Why do we need it?

3 How do we get it?

4 Demonstration of a Business Intelligence Tool

 

 

 

What is Data Warehousing? – David Heise

1. What is it?

The Oracle database used by Banner is designed for OnLine Transaction Processing (OLTP).  A data warehouse is designed for a different purpose.  It is designed to support ad hoc data analysis, inquiry and reporting by end users, without programmers, interactively and online.  This is called OnLine Analytical Processing (OLAP), or Multi-Dimensional Analysis.  It is more than just a better set of reports.  Mostly for performance reasons, a data warehouse is held in a separate database from the operational database, usually on a separate machine.

2. Why do we need it?

The Banner database is designed to be efficient at processing online transactions, and this kind of design is quite inefficient at processing analytical queries and ad hoc reports.  This gives frustrating performance for the person making the inquiry, as well as having a significant impact on the performance of the system for regular Banner users.

 

But there is an even more important reason for building a data warehouse.  The structure of the OLTP database is almost incomprehensible to anyone but a programmer.  Rules of OTLP database design are broken in a data warehouse to make navigation through the subject areas and data elements correspond to business functions, thus making a data warehouse more intuitive to use.

 

This results in at least the possibility of being able to ask questions about the data that can be answered without calling on a programmer.

 

The following table summarizes some of the main benefits that are possible through data warehousing.

 

    1. Has a subject area orientation

    2. Integrates data from multiple, diverse sources

    3. Allows for analysis of data over time

    4. Adds ad hoc reporting and inquiry

    5. Provides analysis capabilities to decision makers

    6. Relieves the development burden on IT

    7. Provides improved performance for complex analytical queries

    8. Relieves processing burden on transaction oriented databases

    9. Allows for a continuous planning process

10. Converts corporate data into strategic information

Table 1. Benefits of Data Warehousing[1]

3. How do we get it?

Some of the key factors to consider when building a data warehouse are listed in the following table:

    1. Be Ready for the Data Warehouse
Develop an understanding amongst senior administrators of the potential role of IT and data warehousing in achieving the institution's goals.

    2. Choose The Right Project Team

    3. Have a Training Strategy
Take appropriate training, and/or hire selected consultants

    4. Choose the Right Architecture
Start small, using a phased approach, but within the framework of a system-wide architecture.

    5. Have a Project “Mission Statement”
Feasibility study
Project Charter
Project Plan

    6. Show Early Business Benefits
Choose strategically important subject areas, (i.e. areas that are linked to the Strategic Plan), that have high visibility and fast return. (remember the 80-20 rule).

    7. Ensure Scalability
Evolve the data marts iteratively, constructing the architected data warehouse as you go.

    8. Understand the Importance of Data Quality

    9. Be Wary Of Vendor Claims
Choose the data repository, data warehousing tools, and desktop tools with care.

10. Use a Proven Data Warehouse Methodology

11. Define and Manage Data Ownership Issues

12. Don’t underestimate the Difficulty of Implementing Change

Table 2 Key Ingredients for Data Warehousing Success[2]

There is a process for choosing strategically important subject areas that starts with the mission statement of the institution, and the goals and objectives.  Critical success factors are then identified, as well as the steps for achieving success as outlined in the strategic plan.  Indicators are found that can be measured and that can be used to monitor the success of the plan, especially in the areas deemed critical, and also adherence to the plan.  These are called key performance indicators, and these form the basis of analytical data models that are provided in the data warehouse.

4. Demonstration of a Business Intelligence Tool

Raw Banner data for 1997 and parts of 1996 and 1998 were imported into PowerPlay, the business intelligence tool from Cognos.  Data reported in the Deans' Statistics report from Institutional Research was imported in this example.  The demonstration is intended to show some of the flexibility in representing and visualizing data, as well as how the data can be “sliced and diced” and viewed at different levels of summarization.

 



[1]   Heise, D.  Data Warehousing at AvondaleCollege. p15
http://dheise.andrews.edu/dw/Avondale/ACDWTOC.html

[2]   Crofts, S.A. You Can Create the Perfect Data Warehouse: Key Ingredients To Data Warehousing Success. (Adapted)  The Data Warehousing Institute.  Data Warehousing Buyer’s Guide. p26