Drawn to Scale — Now your MongoDB can scale to Hadoop size with no performance hit

Drawn to Scale — Now your MongoDB can scale to Hadoop size with no performance hit

The general idea, when speaking about databases, is that SQL powers the big stuff while MongoDB trades off size capabilities for speed. As such, we’ll often see SQL (of some flavor) used in large, enterprise applications while Mongo powers the scrappy startups. San Francisco’s Drawn to Scale thought that the tradeoff was one that nobody should have to make, so the company came up with its own solution, dubbed Spire. It’s a distributed database, written on top of Hadoop, that is custom-built with scaling in mind.

But many companies don’t want to switch away from what they’ve already built, and so today Drawn to Scale is announcing that Spire will now support MongoDB (even with today’s updates) in order to help those scrappy startups scale well into the terabyte range. I had the chance to speak with Drawn to Scale CEO Bradford Stephens and he went into the details a bit more.

“Spire is built from the ground up for really fast, distributed computation. It’s an infrastructure with many different interfaces including SQL and MongoDB. Storage and indexing are all written on intermediary frameworks and languages. The net result is that we provide a universal platform for query languages.”

In an era where everybody loves to talk about big data, nobody seems to address the problems that it can surface. High query capacity, batch-processing in real-time and a lack of efficiency with MapReduce can lead to bigger headaches than what many companies are willing (or able) to address. Switching platforms isn’t always a viable option, so an intermediary solution like Spire can save many sleepless nights and massive expense.

Perhaps the true beauty of what Drawn to Scale has done, however, is in leaving the ability to run SQL on top of the same data. Now, instead of having to choose between the difficult task of sharding out MongoDB or choosing the more robust by slower SQL, Spire gives teams the ability to live in the best of both worlds.

Problems will arise with any system. There have been loads of discussions, blog posts and late nights spent lamenting the choice of one system over another. In a time when the agility of a company has never been more important, tools such as Spire make sure that you can stay in the game when your big data SQL project suddenly needs to serve lightning-fast information as a Web app.

Stephens tells me that the MongoDB support has been in closed beta testing for a few weeks, but Drawn to Scale is looking to roll out the product en masse in the middle of April. Priced at around $500 per server per month, the MongoDB support will be included as a free add-on to existing database customers while it’s still in the beta phases.

Image: Thinkstock

Read next: Square's launch in Canada v. US: 25% higher adoption rate, 90% higher GPV, and 71% higher transaction size