Crate.io, the winner of our Disrupt Europe 2014 Battlefield, is launching version 2.0 of its CrateDB database today. The tool, which is available in both an open source and enterprise version, started out as a general-purpose but highly scalable SQL database. Over time, though, the team found that many of its customers were using the service for managing their machine data and, unsurprisingly, decided to focus its efforts on better supporting those clients.
“The main message is that we hit the nail with the machine data focus that we now have for CrateDB,” Crate co-founder and COO Christian Lutz told me, and added that it was basically the customers who educated the team on what they needed. “We took a look at what they were doing and the database market is so crowded — but we have this mix of SQL and NoSQL,” Lutz said, “and IoT is going to be the fastest growing market for databases.”
So with version 2.0, the team added a number of features that are specifically meant for these kinds of use cases. Thanks to its use of software containers, the overall Crate architecture can already scale automatically to meet demand, but with this new version, this clustering system offers improved performance and resiliency. CrateDB 2.0 also features a couple of improvements to its core SQL engine (aggregation on Joins, sub-selects, table renaming) and new index structures for faster range queries and geospatial queries — the kind of queries you’re very likely to perform on IoT data.
In addition to updating its software, the team also recently launched a managed service business. Companies that want Crate to run their database infrastructure for them can now do so (for a price, of course). To support this service, the team has to run a 24/7 operations team, which is hard to do with the 25 employees it currently has. Chances are then, we’ll see Crate look for funding in the coming months to scale its team as customer demand increases.