DATA WAREHOUSES AND DATA MARTS

If you have been following the literature on data warehouses for the past few years, you

would, no doubt, have come across the terms “data warehouse” and “data mart.” Many

who are new to this paradigm are confused about these terms. Some authors and vendors

use the two terms synonymously. Some make distinctions that are not clear enough. At

this point, it would be worthwhile for us to examine these two terms and take our position.

Writing in a leading trade magazine in 1998, Bill Inmon stated, “The single most important

issue facing the IT manager this year is whether to build the data warehouse first or the data mart first.” This statement is true even today. Let us examine this statement and

take a stand.

Before deciding to build a data warehouse for your organization, you need to ask the

following basic and fundamental questions and address the relevant issues:

• Top-down or bottom-up approach?
• Enterprise-wide or departmental?
• Which first—data warehouse or data mart?
• Build pilot or go with a full-fledged implementation?
• Dependent or independent data marts?

How are They Different?

Let us take a close look at Figure 2-5. Here are the two different basic approaches: (1)

overall data warehouse feeding dependent data marts, and (2) several departmental or lo-

cal data marts combining into a data warehouse. In the first approach, you extract data

from the operational systems; you then transform, clean, integrate, and keep the data in

the data warehouse. So, which approach is best in your case, the top-down or the bottomup

approach? Let us examine these two approaches carefully.

Top-Down Versus Bottom-Up Approach

Top-Down Approach

The advantages of this approach are:

1. A truly corporate effort, an enterprise view of data.
2. Inherently architected—not a union of disparate data marts.
3. Single, central storage of data about the content.
4. Centralized rules and control.
5. May see quick results if implemented with iterations.

The disadvantages are:

1. Takes longer to build even with an iterative method.
2. High exposure/risk to failure.
3. Needs high level of cross-functional skills.
4. High outlay without proof of concept.
5. 
   

Bottom-Up Approach

The advantages of this approach are:

1. Faster and easier implementation of manageable pieces.
2. Favorable return on investment and proof of concept.
3. Less risk of failure.
4. Inherently incremental; can schedule important data marts first.
5. Allows project team to learn and grow.

The disadvantages are:

1. Each data mart has its own narrow view of data.
2. Permeates redundant data in every data mart.
3. Perpetuates inconsistent and irreconcilable data.
4. Proliferates unmanageable interfaces.

In this bottom-up approach, you build your departmental data marts one by one. You

would set a priority scheme to determine which data marts you must build first. The most

severe drawback of this approach is data fragmentation. Each independent data mart will

be blind to the overall requirements of the entire organization.