Mark Smith, CEO and EVP Research, Ventana Research

There seems to be a lot of new marketing around data management software. Is it just that the data ecosystem has grown so quickly?

Organizations have realized that the efficiency of managing their information and the cost of managing your data is a place for improvement and a lot of the top-down pressure is still about cost avoidance. As CIOs focus on the core of their IT budget that is keeping the lights on, three-quarters or more of their budget, there's a need to be more lean and mean about managing data. The other side is that there are more applications that need specific data instances, for Web Internet commerce or other internal projects and organizations need to ask whether the conventional wisdom of using traditional database players the best path or not. The question becomes, "is there something better than what we're currently doing?"

What are some of the hotspots?

There are a bunch of conversations specifically about indexing and the speed of query retrieval and whether the indexes are built properly. Another one is how to parallelize sub-second response times to how data should be appropriately stored and accessed. It's about how you create near-line access to large volumes of data and where should that be managed. A lot of discussion is whether it should be just software, if it should be built into a commodity appliance or should it be a specialized appliance. So you have a range of vendors from Netezza to ParAccel, Kognitio, Greenplum hitting hard on this stuff. A lot of vendors are talking such deep science on these topics that their conversations are only appealing to the top 10 percent of the market. The mainstream data management markets are just trying to know if they should go with a new supplier they don't know much about, and if the cost/benefit ratio over the next three years good enough compared to Oracle, Teradata, Microsoft or IBM.

Aren't the big vendors well advanced in their data management software and tools?

They are, and a lot of times they are not doing a great job of articulating the value of how they can manage data across the enterprise with their distributed management tools. I think they are somewhat in denial that these new plays here are impacting their business. So the conversation around the fastest and cheapest way to manage data for a particular application versus for the entire enterprise is something a lot of organizations haven't really thought about. It's really about multiple instances but a lot of buyers are just trying to solve one problem and make one budget for one application lean and mean.

Wouldn't companies prefer to have a master plan to work against?

Sure, most do if they're well-managed from a CIO data management perspective. But some organizations are not and are more managing distributed data instances, or they're managing one particular data instance and they're less worried about managing across the enterprise. In some cases the CIOs are not good at getting more engaged in the data and information management elements and you find architectures designed just off of one technology.

Isn't that the easiest and most direct route though?

You have to balance these things. If you've been happy managing Oracle and Teradata in the enterprise and now you go out after a third technology, there is some overhead to that. You need to consider supplier management, additional resources, a different technology approach brings a pro and con. If you've made massive investments with other data management suppliers and you're risking a third, if it's going to save you a million bucks a year, just make sure there's not a lot of other overhead that's involved in managing that supplier.

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