March 08, 2012, 9:52 AM — As he surveys the show floor of the Strata conference, Sandy Steier has a look on his face I've seen before.
It's that vaguely bemused-and-tolerant expression New Yorkers get when confronted with something that's not part of their mindset. It's a look that takes in all of the yellow elephants and flashy booths and says "what took you guys so long?"
Steier may have a decent claim on this position: as CEO of 1010data, he and his team have been doing big data for over 12 years. So all the hoopla about big data and data warehousing these days must just seem like the rest of the IT community catching up with what 1010data has been doing all along.
He's quick to make sure the work his company does is well-and-truly separated from the Hadoop messaging machine, too.
"Our involvement in big data is a little different than [Hadoop]," Steier told me firmly. "We're not open source, we provide a proprietary solution. We are NoSQL, and we've been successfully providing our services to over 200 customers for the past 12 years."
1010data is not about Hadoop or clustered data storage, it's about tabular data. A lot of tabular data. The data warehousing firm features hosted data warehouse and business intelligence services for its clientele, with data stored on what 1010data calls its Trillion-Row Spreadsheet.
This is pretty much what it sounds like: a spreadsheet-like interface that stores business oriented data and, though the use of the custom query language Steier and his co-founder Joel Kaplan put together in their Wall Street days in the 90s, delivers a lot of analytical power to customers.
That Wall Street background reveals a lot about the history and structure of 1010data.
"We were our own users," Steier said, "and we missed the whole relational database revolution, because we were building our own stuff."
And when the time came for Kaplan and Steier to move on to other challenges, they looked at the systems they had built and decided that they could be productized to deliver similar results to customers of a hosted service.
1010data's approach makes a lot of sense, especially when dealing with financial- and retail-sector data when nearly everything is tabular in nature anyway. And a spreadsheet is something that nearly any business person worth their salt can wrap their heads around.
The format offers more than just a familiar interface for users.
"With 1010data, we have a highly interactive interface with data," explained Tim Negris, VP of Marketing. "Hadoop and SQL-based systems are batch oriented." Which leads to a less real-time experience in analytics, he added.
Because the entire solution is hosted, it also reduces the amount of overhead the customer has to implement. The interface is also geared to quickly enable searches and analysis on these massive-scale tables, which can get up to 4 petabytes in size.
This is not, clearly, a solution for everyone. Unstructured data doesn't fit in this kind of DBMS, and even structured data, such as e-mails or logs, could be difficult to fit in the tabular structure. But given that so many enterprises still store their data in spreadsheets and RDBMS tables, that's still a lot of potential customers out there for the 1010data crew.
"We have an alternative solution and we've had it for a long time," Steier emphasized. "We are an important alternative among the alternatives."
Read more of Brian Proffitt's Zettatag and Open for Discussion blogs and follow the latest IT news at ITworld. Drop Brian a line or follow Brian on Twitter at @TheTechScribe. For the latest IT news, analysis and how-tos, follow ITworld on Twitter and Facebook.