Atlas Mapped: New Cross-Region VPC Peering And 60 Regions

The MongoDB Atlas platform is constantly evolving and in this new, regular column, we're going to summarize what's launched in Atlas over the past few weeks to get you up to speed with its rapid evolution.

  • Atlas Cross-Region VPC Peering on AWS
  • Atlas launched in AWS Stockholm
  • Control Your Maintenance Times
  • Custom MongoDB Roles for Atlas
  • Workload isolation for analytics

Cross-Region VPC Peering on Amazon Web Services

First up is Cross-Region VPC Peering. Before this arrived you could connect your Atlas instance's VPC to your application's VPC as long as they were in the same region. But not everyone's applications are that logically close to where they've chosen to site MongoDB. Now with the Cross-Region VPC Peering support, you can have your application and database in different regions. Read about how you can set up VPC Peering in the documentation, and do remember this is only for M10 and larger dedicated instances. The non-dedicated M0, M2, and M5 database instances do not support VPC Peering.

Stockholm AWS takes Atlas to 60 regions.

We've just passed a landmark in Atlas's expansion. We've just enabled Atlas in Amazon's Stockholm datacenter which has taken the combined total of regions (on AWS, GCP, and Azure) that Atlas is available in up to the grand total of 60. Of course, the new data center is also great news for Scandinavia. This is the second Nordic region we've opened up; the first was in Finland on GCP. Stockholm is also the first home of the M700 instance. With 96 vCPUs and 768GB of RAM and 4TB of storage, it's a RAM-rich beast of an instance.

Control Your Maintenance Times

Atlas will, by default, do routine maintenance on a weekly basis at its own selected time. With this new "Preferred Cluster Maintenance Start Time" in the project settings, you can take control of what time that maintenance starts. We also let you select what time-zone that time is so you don't have to do all the "timey-wimey" stuff in your head. That makes it easier than ever to avoid your peak traffic times for that essential work that keeps your clusters running up to date and smoothly.

Custom MongoDB Roles for Atlas

If you've got a scenario where you need a special role for a user or users and you can't match it up with built-in Atlas MongoDB roles then this option is for you. Custom MongoDB Roles can now be created and applied to users through the Atlas UI or API. You can pick any privileges or inherit them from another role to create your new roles and get your access control model to fit your people just right. Find it in the Clusters View, under the Security Tab, marked MongoDB Roles.

Workload Isolation for Analytics

Analytics can be punishing on your production workloads - but not anymore. Now on Atlas, you can enable separate analytics nodes in your cluster which can be dedicated to taking the load of those analytic queries, isolating them from your operational workloads. That keeps your operational workload running at peak performance while letting you dig deep for the analytics insights you need.

You can find this new feature in the Atlas cluster builder where you currently configure multi-region and replication options. It's simple to target analytics nodes; just use the "Analytics" tag when setting your read preference. If you have the BI Connector for Atlas enabled, which allows you to connect your managed database to any SQL-based visualization tool, the connector will automatically default to using your analytics nodes.

That's it for this edition of Atlas Mapped. Look out for future editions as we keep you right up to date.