Channels

- For Consultants
- For Developers
- For Managers
- For Professors
- For Researchers
- For Students


Multidimensional Analysis Brief

In an interview at dbazine.com, Ralph Kimball states "The dimensional approach is a very simple symmetrical approach thatís organized around taking measurements. I wonít go on at length about this, but if you think of computers as taking measurements of things, like, how much did the customer buy today or what was the charge that we made, then any transaction thatís made in any business can be thought of as a measurement. We put those measurements in at the core of our databases, and those end up being what I call fact tables in the dimensional approach. Then you basically surround those measurements, which are very physical, you just surround them with what you know, which is OK, we know that this was the customer, we know that this was the store, we know that this was the time, and those end up being the dimensions. So thatís sort of the whole thing ó itís a very natural, simple approach.

According to an IBM web tutorial "Tools that employ OLAP technology, such as the DB2 OLAP Starter Kit, empower users to ask intuitive and complex ad hoc questions about their business, such as, "What is my profitability for the third quarter across the southeast region for my focus products?" Such a question requires multiple perspectives on the data, such as time, regions, and products. Each of these perspectives are called dimensions."

"Relational data can be considered two-dimensional because each piece of data, which you can also call a fact, correlates to one row and one column, each of which can be considered a dimension. The dimensions in a multidimensional database are higher-level perspectives of the data that represents the core components of your business plan, such as Accounts, Time, Products, and Markets. In an OLAP application, these dimensions tend not to change over time."

"Each dimension has individual components called members. For example, the quarters of the year can be members of the Time dimension, and individual products can be members of the Products dimension. You can have hierarchies of members in dimensions, such as months within the quarters of the Time dimension. Members tend to change over time, for example, as your business grows and new products and customers are added."

  • Concepts of Dimensional Modeling
  • IBM Multidimensional Analysis Tutorial page
  • Interview with Dr. Ralph Kimball by C. Sarkar.