In-Depth

Netezza Announces Massive Data Warehouse Appliances

Netezza fleshed out its line of with new 50 and 100 TB offerings; the latter is nearly quadruple the size of its existing high-end product

• By Stephen Swoyer
• 05/25/2005

This month, Netezza Corp. fleshed out its line of data warehousing appliances with new 50 and 100 TB products.

Netezza’s new NPS 10000 Series includes two models, the NPS 10400 and NPS 10800, which are designed to support data warehouse footprints of 50 and 100 terabytes, respectively.

It’s an aggressive expansion for Framingham, Mass.-based Netezza, whose high-end 8000 Series was previously capped at 27 TB.

What’s more, says VP of marketing Ellen Rubin, Netezza isn’t ahead of the market, either. She notes that customers in several verticals have requested appliances of 50 TB or more.

“This is where we see the market heading. We have customers—large telcos, companies that have a lot of clickstream data—saying, ‘We need headroom to grow,’” she says, noting that some customers are experiencing 100 percent year-over-year data warehouse growth. “We’re also moving into the federal market, and there’s definitely demand there, as well, for these larger systems.”

In tandem with its NPS 10000 Series launch, Netezza also announced a new version 3.0 release of its NPS appliance operating environment. NPS 3.0 supports several long-awaited goodies—such as accelerated query processing and bulk data movement capabilities—but Rubin says there was no inherent size limitation, per se, in previous versions of NPS.

“There was definitely [a limitation] at the beginning—when we first announced the product, we were dealing with data sizes that were up to 10 TB, but that was, again, mostly determined by the market. Then we quickly saw demand outstrip that set of configurations, so we added an additional set of configurations that took us up to 27 [TB]. I think for the moment that 100 [TB] ought to keep [pace with demand] for a while, however.”

Netezza has always touted ease of use as one of the biggest advantages of the data-warehouse appliance model. Now that the company proposes to nearly quadruple the maximum size of its flagship appliance, do the same ease-of-use advantages—particularly with respect to installation, population, and ongoing administration of the data warehouse—still remain?

Rubin, not surprisingly, says they do.

“That was one of the big questions [at Netezza’s recent user conference]—how as you’re moving up into these enormous data sizes can you maintain the ease of use and simplicity? But one thing to keep in mind is that the footprint of these two [10000 Series] products isn’t much bigger than it is today,” she says. “In terms of the manageability, you compare traditional data warehouse systems where you can have football fields worth of equipment, here we’re dealing with very, very small footprints—literally eight rack [units] for 100 TB.”

Netezza and high-end data warehousing specialist Teradata, a division of NCR Corp., have an interesting relationship. They disagree about a great many things—e.g., Teradata says Netezza is a data mart appliance vendor, Netezza (understandably) says it’s a lot more than that—but they have reached a rare consensus on one issue: namely, both vendors claim that they don’t actually compete all that much against one another (see links to related articles, below).

With the launch of Netezza’s 10000 Series appliances, however, this issue seems moot: like it or not, Netezza is vying for share in the same 50 TB-and-above stratosphere that—by virtue of defining the high-end of the high-end data warehousing space—is also home to Teradata and a handful of other vendors. Rubin, for her part, effectively concedes the point.

“We consistently see three major players up in this very-large-size data warehouse market—Oracle, IBM, and Teradata,” she says. “But I don’t think it’s accurate to say that it’s going to position us more against Teradata than in the past. This gives us the capability to sort of help our customers move Netezza into more of an enterprise-level type of environment. Where in the past our enterprise data warehouse might scale beyond tens of terabytes, this opens up the headroom for us to continue to scale in that direction.” And what of Teradata’s claim that Netezza is an application-specific data mart play? To some extent, Rubin concedes, this was true—at least three or four years ago, when the company was still a start-up. “The reality actually is that our customer base is usually split. When we started, it was more typical for us to be considered for an application-specific [data mart], because—think about it—who’s going to trust a new vendor?” she says. “But obviously at this point, when we have close to 30 customers, it gets more credible and believable to say that we are going to be able to handle a much more strategic role in the data center.”

Related Articles:

Q&A: Managing Exploding Data Volumes
http://www.esj.com/business_intelligence/article.aspx?EditorialsID=7476

Data Warehousing Appliances: The Next High-End Trend
http://www.esj.com/business_intelligence/article.aspx?EditorialsID=7452