Data Warehousing: Tips, Trends and Resources

November 7, 1996

If you rarely accumulate frequent flyer miles to the destination named data warehousing, then check your passport and tighten your seat belt. Get ready for a litany of lingo, vintage acronyms and linguistic wrinkles you will need to navigate this newest market segment. You come across terms such as EIS (executive information system), DSS (decision support system), data mart, data mining, OLTP (online transaction processing), OLAP (online analytical processing), ROLAP (relational OLAP), busines

For systems integrators and data managers alike who seek data warehouse immersion, there's not only aid and comfort, but a list of guidelines to keep you on the straight and narrow path as you build up stamina for the steep climb ahead. Data warehousing is here to stay and likely will remain a key focus of business users for quite some time.

If you rarely accumulate frequent flyer miles to the destination named data warehousing, then check your passport and tighten your seat belt. Get ready for a litany of lingo, vintage acronyms and linguistic wrinkles you will need to navigate this newest market segment. You come across terms such as EIS (executive information system), DSS (decision support system), data mart, data mining, OLTP (online transaction processing), OLAP (online analytical processing), ROLAP (relational OLAP), business intelligence, decision groupware and more.

Your First Stop

The first stop on your journey should be the Data Warehousing Institute, Bethesda, Md. (http:// www.tekptnr.com/tpi/tdwi). Its World Wide Web site proclaims it the "authoritative compilation of data warehousing tools, processes, case studies, white papers, articles, publication lists and conference schedules, plus news about recent developments in data warehousing." Surprisingly for the Internet, that's pretty accurate.

One of the more valuable bits of wisdom on the site for anyone taking on the task of creating and maintaining a data warehouse is the work of Alan Paller, TDWI's director of research. Paller compiled a list of "Ten Mistakes to Avoid for Data Warehousing Managers" from industry experts and through regional meetings hosted by the institute. They include the following:

-Starting with the wrong sponsorship chain

-Setting expectations that you cannot meet and frustrating executives at the moment of truth

-Engaging in politically naive behavior, for example, saying "This Will Help Managers Make Better Decisions"

-Loading the warehouse with information "Just Because It Was Available"

-Believing that data warehousing database design is the same as transactional database design

-Choosing a data warehousing manager who is technology-oriented rather than user-oriented

-Focusing on traditional, internal record-oriented data and ignoring the potential value of external data and of text, images, and -- potentially -- sound and video.

Other mistakes to avoid, he said, are: delivering data with overlapping and confusing definitions; believing the performance, capacity and scalability promises; and, believing that once the data warehouse is up and running, your problems are finished.

And, what, you ask, is the No. 1 mistake?

"Why, [the first one], of course. The driver for the data warehouse, that is, the person between the sponsor and the data warehouse manager, can't be IS," says Paller. "The sponsor is at the top, the executive with a great deal of money to invest in the effective use of information. The driver is the one who keeps the project moving in the right direction and makes sure the schedule is kept. A good driver is a business person with three essential characteristics: (1) already earned the respect of the other executives, (2) has a healthy skepticism about technology and (3) is decisive but flexible."

Your Second Stop

For your second stop on the data warehouse tour, drop by Larry Greenfield's Data Warehousing Information Center (http://pwp.starnetinc.com/larryg/). The financial and marketing decision support systems consultant for LGI Systems Inc., Northfield, Ill., created and maintains this guide, which includes vendors of data warehousing and decision support technology.

"I tried to present a clear overview of what's available in data warehousing," says Greenfield. "The pages should help those looking for end-user tools and commercial sites. For the novice, check the list of activities named, 'Getting Started with Data Warehousing.' You can find it under the category, Rants and Ravings."

In his no-nonsense, no-frills layout, you discover a wealth of references and resources: vendors of report and query software, OLAP/multidimensional data-bases, document retrieval systems, data extraction, cleaning and loading software, information catalogs and relational databases for data warehousing. There's also a neatly arrayed table of data mining tools, which covers the name of the tool, the company responsible, its phone number and a brief description. The tools are separated into the following categories, which Greenfield admits may overlap: case-based reasoning, data visualization, fuzzy query and analysis, knowledge discovery and neural networks.

If that's not enough, you can even gather articles, white papers, technical evaluations and lists of organizations for training and conferences, associations, periodicals, newsgroups, mailing lists and Web sites.

Your Third Stop

And where precisely is this segment heading? In true data warehousing fashion, the Data Warehousing Institute gathered 25 members and together created a consensus document suitably titled, "Top Six Trends in Data Warehousing for 1996."

According to the experts, six major trends will drive the surge in data warehousing implementation:

1. World Wide Web browsers will access data warehousing applications

2. Relational database management systems will get performance boosts

3. Relational OLAP and traditional OLAP will both grow rapidly as tools to analyze data from the warehouse

4. Users will 'fess up' and clean up their bad data

5. Data mining will explore data warehouse information

6. OLTP and Data Warehousing will team up

The Rest Stop

For the systems integrator who wants to delve even more deeply, he or she will find detailed information on these pages:

-Corporate KDD Bookmark: http://www.cs.su.oz.au/~thierry/ckdd.html

-Data Warehousing Information Center: http://pwp.starnetinc.com/larryg/

-Data Warehousing Institute: http://www.tekptnr.com/tpi/tdwi/