TDWI | Training & Research | Business Intelligence, Analytics, Big Data, Data Warehousing

Teradata Goes for Appliances in a Big Way

Teradata explains the importance of appliances -- as both data warehouses and data marts -- and trumpets its Active EDW vision.

• By Stephen Swoyer
• March 11, 2009

It's been almost a year since Teradata Corp. surprised the data warehousing (DW) industry by announcing its first branded DW appliance systems. (read more here).

That was just beginning. In September and October, Teradata unveiled several additional appliance systems, including a new Extreme Data appliance that -- at $16,500 per TB -- seemed priced to move (see http://www.tdwi.org/News/display.aspx?ID=9197).

Teradata's about-face was complete: after years of downplaying the significance of DW appliances -- likening them to band-aid solutions that are tapped primarily for data marts -- the company is now promoting them.

"For the most part, appliances were getting lot of play as a cheaper thing, a faster thing. If you put the name 'appliance' by your product, it was [regarded as] a good cheap thing," says Randy Lea, vice president of product and services marketing with Teradata. "Our whole position was, 'What is an appliance?' There were a lot of definitions. … There was a feeling [that] if we called our premium product [Teradata's Active Data Warehousing platform] an appliance, we would be positioning ourselves with these other, quite frankly inferior, products."

So Teradata shifted gears, Lea says, emphasizing platform flexibility over its traditional one-platform strategy. The only thing missing, of course, was a stratified product line. Over the course of 2008, Teradata fleshed out a product line-up that now includes data marts, DW appliances, highly-scalable DW appliances (its Extreme Data systems can scale up to 50 petabytes), and its high-end, Active Data Warehousing systems.

"Our platform family is targeted for different purposes. One of those, the data warehouse appliance, is positioned against the other appliances on the market," Lea explains. "Quite frankly, we don't call our Active Data Warehousing platform an appliance, but the only reason for that is that we couldn't have. It is configurable for different performance levels, so it's not quite plug-and-play. [With] our Data warehouse appliance, our data mart appliance, [and] our Extreme Data appliance, there really are no options: you plug them in and they're ready to go."

What's more, the Teradata of today is clearly more comfortable using the term "appliance" that the Teradata of old simply would not (under any circumstances) have countenanced. Lea concedes the point.

"We are more open to positioning ourselves as an appliance. We're actually using it in the naming of the products. We are an appliance company. We always have been an appliance company," Lea says. Maybe, but Teradata certainly didn't tout its appliance bona-fides. Quite the contrary, in fact.

Why the sudden about-face? Lea returns to the issue of dilution: terming Teradata's then-premium Active Data Warehouse systems "appliances" would have invited comparison with (and would have elevated the stature of) the DW appliance upstarts. (In the past, officials frequently cited both IBM Corp. and Oracle Corp. -- and not Netezza Corp. or, more recently, Dataupia Inc. -- as Teradata's biggest competitors.) So Teradata continued to champion its inflexible enterprise data warehouse (EDW) vision, which -- Lea now concedes -- wasn't the best fit for every customer.

"There are these splintered [groups inside of] organizations that will build a particular data mart … because it's their data … [so] they want to control it, or maybe for political reasons. Even for customers [who] are executing an enterprise data warehouse, there are data marts being built. We never have really gone after that business, never really focused on that business, and, quite frankly, we priced ourselves out [of some opportunities] because of that," he comments. "With our family of products, price is not really an issue. We're not trying to take our Active Enterprise Data Warehousing platform and trying to squeeze it into a certain price point. We now have price points across the board."

The attraction, he claims, is that customers can buy into Teradata at several different levels -- starting at $16,500 per-TB.

Perhaps, but Teradata's appliances can't be upgraded without ripping and replacing them to its high-end Active Data Warehousing configurations. Teradata's data mart and data warehouse appliances can be repurposed for test or development purposes, or otherwise tapped as back-up nodes for failover or fault-tolerance. However, when a customer is ready to step up to Active Data Warehousing, it must do so on new hardware.

Lea, for his part, downplays this as a potential drawback. Hardware costs are negligible, he contends, and Teradata's value proposition is portability up and down its product line -- from its entry-level data mart appliances on up to its flagship Active Data Warehouse systems.

"What we mean by that is [that] the software is the same, so all of your ETL processes, all of your data models, all of your application connectivity, all of the testing, all of the other things around the platform are all portable," he comments.

"You can't take the data warehouse appliance and attach it to our active enterprise data warehouse. What customers are doing is taking those platforms and using them in test and development systems," Lea continues, circling back to the question of hardware pricing. "One of the things we got from CIOs is … they're telling us, 'The actual cost of my training, my ETL, my application development -- all of that can be saved. That is probably 2-to-1 over the hardware purchase, so if I can leverage all of that, … that is huge."

An Enterprise Data Warehouse in Every Pot?

Teradata seems to have moderated its positions in a few other respects, too.

In the past, for example, Teradata was an ardent champion of the enterprise data warehouse, prescribing an EDW for each and every enterprise -- no exceptions.

Last year, Lea and Teradata refined that message, backing away from the idea that an EDW is the only solution and embracing pragmatism: an EDW might not be the only solution, although it's still the best of all possible solutions. Nevertheless, data marts are sometimes needed, a position the new Teradata is at last willing to countenance (see http://www.tdwi.org/News/display.aspx?id=8968).

One year on, Lea revisits this issue, stressing that Teradata's concession to data marts in no way amounts to an abandonment of its EDW vision.

"We frequently get asked, 'Wait a minute, appliances are used for data marts. Does that mean you're no longer advocating your enterprise data warehouse -- your integrated data warehouse view?' No, it doesn't. We've always believed … that the value of integrated data drives differentiated answers and questions, differentiated actions that you can take against your competitors for competitive advantage," he maintains. "We are always going to position our enterprise data warehouse as the primary architecture from a business perspective."