Does the world really need yet another data warehouse technology—let alone one that defies convention and flaunts compliance concerns by living outside the “safety” of corporate data centers as part of a public cloud service?
Snowflake Computing, a startup in San Mateo, Calif., founded by a team of data experts including two Oracle engineers and helmed by longtime Microsoft (MSFT) executive Bob Muglia, has raised $26 million to answer that question.
Emerging from the shadows this week with backing from investors including Redpoint Ventures, Sutter Hill Ventures, and Wing Ventures, Snowflake’s proposition is simple. It believes its patent-pending SQL relational database technology—built completely from scratch—can accommodate more types of business data than legacy offerings from the likes of Teradata (TDC) or big data management platforms building on Hadoop. “Customers do have existing solutions, but they’re not satisfied. It’s not like we’re entering a market where people are happy with their current solutions,” Muglia says.
Rather than requiring customers to install its technology on site, Snowflake is offering it as a service. Businesses pay for how much data they are storing and by the number of hours it takes to analytics queries against their information. The most direct competitor is Redshift from Amazon Web Services
. The company claims its approach costs 90% less than investing in the hardware and software necessary to build on on-site data warehouse.
Snowflake was founded two years ago by a team of engineers who between them hold more than 120 patents in databases and data management technologies: Oracle (ORCL) veterans Benoit Dageville and Thierry Cruanes, and Dutch computer scientist Marcin Zukowski. The whimsical name pays homage to their mutual love of snow sports. Muglia joined as CEO in June, taking over from Sutter Hill director Mike Speiser, who was managing the team while it was in stealth mode.
Muglia admits Snowflake won’t appeal to companies that aren’t willing to put their data on cloud servers and storage. Its initial customers hail from the advertising, media, and technology sectors.
“Whoever has the biggest dataset can answer the hardest questions,” says James Rooney, senior vice president of media platforms at Accordant Media, one of Snowflake’s beta customers. “Instead of spending an hour waiting for a response, we get it in five minutes with Snowflake. So we can spend more time interpreting the result or whiteboarding harder questions.”
Whether Snowflake can melt the hearts of skeptics remains to be seen. Still, if you consider how much data is now being generated in applications and services that live “in the cloud” (as opposed to on-premise corporate data centers) the logic behind its approach could add up quickly.
This item first appeared in the Oct. 21 edition of Data Sheet, Fortune’s daily newsletter on the business of technology. Sign up here.