MongoDB has expanded its Atlas cloud database service to enable multi-cloud operations across AWS, Google Cloud, and Microsoft Azure.
The new MongoDB Atlas multi-cloud cluster capability became generally available on Tuesday and expands on the company’s existing cloud database offering. MongoDB Atlas first launched in 2016 as a cloud database service running on AWS, and expanded to support Microsoft Azure and Google Cloud in 2017.
The new multi-cloud cluster support isn’t about simply running a managed MongoDB database on different cloud providers. Rather, with the new update, MongoDB allows users to run a database across multiple cloud providers at the same time, which was not easily possible prior to the update.
“As a popular database choice for many organizations, it makes a lot of sense for MongoDB to go multi-cloud with Atlas, simply because their customers are asking for it and if they aren’t asking for it now, they will in the future,” said James Curtis, senior research analyst at 451 Research, part of S&P Global Market Intelligence.
Curtis noted, however, that multi-cloud can mean different things to different organizations. In his view, the reality is that multi-cloud best serves targeted workloads such as high availability, backup and specialized analytics, which MongoDB has highlighted with this Atlas update.
Running MongoDB Atlas as a multi-cloud database
Generally, MongoDB customers that are multi cloud run separate applications across AWS, Azure and Google Cloud, said Sahir Azam, chief product officer at MongoDB.
While that has been the case, Azam said there has been demand from organizations for multi-cloud data portability. The growing use of container technology by organizations has made it easier for application code to be portable, he said, but the core data and databases used by applications have remained tethered to one cloud provider. The basic idea behind the new multi-cloud clusters is that a database application can run across multiple cloud providers and regions within those providers.
How MongoDB Atlas multi-cloud clusters work
MongoDB at its core is a distributed database and has a replication protocol that can keep the distributed nodes of a particular database in sync. In the past, that replication protocol was used to keep local databases on-premises, or a particular cloud provider in sync. With multi-cloud clusters MongoDB Atlas now provides live replication across cloud providers, Azam explained.
This connects the security and networking in a seamless way across those clouds, he said.
For example, if an organization needs to keep all its data within Germany for compliance reasons, it can benefit from using multiple regions across AWS, GCP and Azure. Azam noted that typically no major cloud provider today typically has more than one region in that country and so now, they can actually have the ability to span multiple regions within Germany.
Data queries also benefit from the distributed MongoDB Atlas multi-cloud cluster approach. Azam said that within the core MongoDB database itself is the ability to route a query to the closest node across a distributed deployment. With multi-cloud clusters, users now will have the ability to query the closest data node, regardless of cloud provider, to get a faster response.
“Multi-cloud effectively just gives customers more reach,” Azam said.