The Coming Data Battles


Every now and again you run into somebody at a company that just seems to get it a little bit more than everybody else. In the case of Veritas, that person may very well be Matt Fairbanks, the company's director of technical marketing.

Over the last two years, Veritas has acquired a number of companies that has moved the company beyond its core backup business to embrace data and application management and even server provisioning.

As part of that strategy, Veritas has loosely positioned these various products and technologies as a suite of tools for enabling utility computing across multiple platforms.

But behind that me-too buzzword approach to the market is an evolving next generation approach to data management that is worth taking some time to understand. Ultimately, what Veritas is trying to accomplish is an ambitious effort to create a service-oriented architecture for data. Fairbanks seems to be one of the few people at Veritas that can succinctly articulate a strategy that calls for Veritas products to eventually touch all data in the enterprise, regardless of whether that data resides in storage or in an application running on a client or a server.

With that higher level of virtual abstraction of enterprise data established, Veritas can then develop a range of policy-based data management tools that can be applied across the enterprise. In effect, this will allow IT organizations to create a policy for how all data related to a particular business process, such as accounting, is going to be managed.
Given the overriding interest in compliance and business intelligence in general, the ability to do that would give Veritas a significant differentiation in the market.
Of course, IBM, Oracle, EMC, BEA have similar ideas, but each of them is coming is approaching next generation data management from the point of view of a particular platform. IBM, for instance, sees a world where all data is accessed through Websphere, while Oracle still sees the database as the center of the universe.

What's intriguing about the Veritas approach is that there is no center, which more closely mirrors what is happening in corporate environments that have data distributed across databases, file systems and applications that will eventually be loosely coupled using XML Web services.

Whether Veritas can really articulate and then execute this strategy is anybody's guess. Both EMC and Computer Associates are going to do their level best to cut off the revenue stream that Veritas gets from its backup business. And under CEO-elect John Swainson, CA is likely to fill out its own robust data management strategy to compete more aggressively.

And even more ominous is the probability that all this convergence in the data management space is going to lead to some major mergers and acquisitions as diverse categories such as storage, data and application management continue to converge with the database, application server, portal and identity management space. In short, this pending convergence is going to be the next big war in the enterprise

But in the short term, all this convergence is more than likely going to confuse the market, rather than enlighten it. So that's why people like Fairbanks may soon be in high demand because while there are a lot of people who conceptually understand what is happening, the number of people who can make sense of it are very few and far between. And as in any war, the side with the shortest lines of communication usually wins.