Once upon a long cyber time ago, some folks working for IBM conceived of the relational database and incarnated this concept into a database management system (DBMS) called DB2, née "System R." IBM and most of its customers tended to view DB2 as a "gee whiz" product rather than a serious DBMS player. Besides, IBM already offered an alphabet soup of data management products (IMS, VSAM, ISAM, CICS, etc.); and in the days of low-performance, expensive processors, DB2 (which consumed a lot of cycles) didn't appear to be a very attractive alternative.

Then along came an IBM salesman who believed in relational DBMSs. He suggested that IBM put major eggs in its DB2 basket. When IBM chose not to go along with his recommendation, he started his own relational DBMS company and in short order accumulated a personal net worth of some $8 billion. Along the way, his company, Oracle, did some major butt-kicking, the primary recipient of which was--you guessed it--IBM.

It turns out that IBM, like Rip van Winkle, was only asleep for twenty years--not dead. Like the Phoenix rising from the ashes, DB2 has been reincarnated into a modern, efficient, reasonably priced product that can compete head-on with anything else in the market.

In 1997, IBM introduced DB2 Universal Database. It runs on every popular platform except the mainframe (coming next year). It contains all the bells and whistles touted by its competitors including support for all data types (text, pictures, video, etc.) in addition to relational data; it scales from the desktop to large-scale, enterprise-level systems; it provides users an "extender" mechanism for tailoring to custom requirements a la blades and cartridges; and it has the robustness and reliability one would expect from a company that specializes in such features.

In addition, DB2 Universal Database has some features which will no doubt stimulate the competition. First, DB2 is full-function. There are no "lite" versions, a feature that can be very important to the notion of scalability. In effect, the same system runs (or soon will) on an NT-based uniprocessor workstation that runs on a Solaris cluster that runs on an SMP massively parallel processing system that runs on a mainframe! Second, DB2 has more features that make it "Web-enabled" than any other system. Web-enablement, coupled with the ability to access data from any platform, even DOS and MacOS, should give IBM a powerful tool for competing in the market for enterprise-wide information distribution and collection.

Putting its money where its mouth is, IBM spent more than seven times the amount of money on DB2 promotion in 1997 than it did in 1996, and this budget is expected to increase again this year when much of the promotional focus will be on data warehousing/decision support (which IBM calls business intelligence). In the past, IBM tended to promote associated tools, features, functions, alliances, etc., independently. In 1998, the company promises that its customers and prospects will hear a story that articulates DB2 as the centerpiece of a solutions-oriented business intelligence strategy, i.e., "DB2 everywhere."

IBM also plans to make significant investments in DB2 to support its business intelligence initiatives: for example, the inclusion of new operators to provide OLAP functionality within the database engine itself; support for star schema; built-in tools for data mining, scrubbing and extraction functionality; and middleware that provides for distribution of corporate data via extranets.

With the future viability of Informix and Sybase in question, only NCR, Oracle, Red Brick, Tandem/Compaq and possibly Microsoft will be left to compete with IBM for enterprise-level decision support and information distribution systems. Given the substantial marketing and development resources that IBM is and will be investing in DB2, coupled with the investments it will make in the companion programs associated with its business intelligence solutions unit, we expect that DB2 will emerge as one of a very small number of winners.

Ironically, DB2 could be the product that allows IBM to end-run Microsoft. While Microsoft's SQL Server on NT will no doubt be a big unit seller, DB2 on NT can scale far beyond the Microsoft product and can offer platform independence, including mainframes. Large enterprises, especially those with blue shops, will no doubt feel quite comfortable with a DB2-based solution.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access