MongoDB enters the Database as a Service market with Atlas

mongodb ceo

MongoDB’s popular NoSQL database is now available on-premise, in the cloud and as a service

Share

Open source NoSQL database MongoDB has launched its popular database as a managed service in the cloud on a pay-per-hour basis called Atlas. 

Announcing the product at the MongoDB World user conference in New York yesterday, Eliot Horowitz, CTO and co­founder of MongoDB said: “The day we founded MongoDB, we envisioned a world where developers could focus on what’s most important, turning their giant ideas into innovative applications.

"By running on MongoDB Atlas, developers can trust that their applications will be highly available, secure, scale seamlessly, and without the downtime that comes from applying patch updates.”

Atlas is Database as a Service (DBaaS), so it takes care of operational tasks such as provisioning, configuration, patching, upgrades, backups, and failure recovery, allowing developers to concentrate on their applications. 

A recent survey of more than 2,000 members of the MongoDB community found that 30 percent of respondents deploy on more than one public cloud, so Atlas is designed to operate across public clouds including their partners AWS, Azure and Google.

Read next: MongoDB to challenge Oracle's enterprise dominance, says 10gen CTO

MongoDB CEO Dev Ittycheria said: “This release is yet another major milestone in building the best database for modern applications, and providing expansive options for how customers consume the technology – in their own data centres, in the cloud, and now as a service.”

Like Uber for databases

Naturally handing over these operational considerations to a vendor comes at a premium. Kelly Stirman, VP of strategy at MongoDB explained it as the difference between owning a car against using Uber.

“In some cases it is better that I own a car as I have particular needs, we are a family of seven so we need the space, and we ski so we need the roof rack. So I want all of that control but I have to take responsibility of the car, I have to put gas in it, insure it,” Stirman explained.

With Uber all of these concerns, and the overall control, are handed over to the service provider, and you pay a premium for that convenience factor. “So Atlas is the Uber -  click a few buttons and we install it, configure it, optimise it and secure it,” Stirman said. 

Read next: YouGov creates unified global database with MongoDB

This fits into MongoDB’s overall strategy of providing a database solution for a range of customer needs. “We have offerings across the board and we aren’t locking you in to one or the other. If we don’t provide value you can take it and run it however you like,” Stirman said.

How does it work?

VP of cloud engineering Cailin Nelson in her demo of Atlas showed how customers can build a new cluster in under ten minutes. First you give it a name and a location, with Atlas available immediately across AWS data centres, with support for Microsoft Azure and Google Cloud Platform coming later.

Then you choose an instance size, the desired amount of RAM, storage, capacity and speed. You can opt for encrypted volumes and choose your replication factor from three to five to seven nodes. For larger apps you can opt to run a sharded cluster and select backups.

“Atlas scales with your application” Nelson said, so if you need more disk space you simply access your cluster and change the parameters.

While you cycle through these options the pricing will change on the right hand side before you decide to launch the cluster. Pricing starts at $0.026 per hour for 2 GB of memory and 10 GB of storage, depending on which region you use and scales up depending on instance sizes, storage performance, and backup schedules.

Analyst perspective

Forrester analyst and author of its Database-as-a-Service market overview Noel Yuhanna told Computerworld UK that this move to DaaS was inevitable from MongoDB and that “having a common framework, being able to scale as much as on-premise on the cloud, tooling and compatibility will be critical.”

When asked if he thought this would be a good financial move for MongoDB, Yuhanna said: “It will have a big impact on income. Not overnight, but over the next few years.” 

Carl Olofson, research vice president at IDC said: “This is critical for [MongoDB]. If you’re a database vendor and you don’t have a Database-as-a-Service offering then you aren’t fully engaged.”

Find your next job with computerworld UK jobs