Drawn to Scale, a two-year-old Seattle startup developing a new database called Spire, has raised a round of funding from RTP Ventures, IA Ventures, and SK Ventures. Terms of the deal were not disclosed in a blog post from co-founder Bradford Stephens, a former engineer at Visible Technologies and Microsoft.
“Spire is, for me, the culmination of years of thinking about how a database should be, not how it’s traditionally been,” writes Stephens. “We’ve all dealt with the frustration of keeping an RDBMS alive as it changes and grows. It seems that it only takes a minor incident (too much data, need to change a column, too many users) in order to bring the DB down and get a call at 3AM. We’re changing that.”
The Spire database, currently in private beta, is designed to scale as a service grows.
“Unlike sharded databases, where you have to query every machine and aggregate the results, Spire knows what data your queries need,” the company writes on its Web site. “This means you can handle far more data and users with less latency.”
At the GeekWire Summit on Wednesday, former Microsoft chief software architect Ray Ozzie singled out database technology as one of the areas that he’s most excited about right now.
“I know this is boring, but I’m going to take it out of the consumer device realm, because we’re all enamored with consumer devices, but I am fascinated by the development of NoSQL databases, and some new new SQL databases,” Ozzie told the crowd of more than 500.
He cited the work of Mike Stonebreaker in Boston and Jim Starkey at NuoDB, a relational database.
“The database technology is really fascinating the way it’s developing and maturing right now,” he said. “That kind of has a lot of my focus.”
Other members of the Drawn to Scale team include Ryan Rawson, who previously worked at Amazon.com, Google and Stumble Upon, as well as Alex Newman, a former architect at Cloudera.
For years, we’ve worked at some of the first “web scale” companies — Google, Amazon, Intel, Microsoft, etc.,” Stephens tells GeekWire. “One thing we noticed is how much pain is caused by the traditional database. When something breaks, it’s always the DB. When you can’t roll out a feature, it’s always the DB. And the most expensive piece of software to maintain is the DB. So we set out to build a database that’s easy to use and stays ‘real-time’ no matter how much data or how many users you have.”