Atlas Search Nodes: Now with Multi-Region Availability

At MongoDB, we are continually refining our products to try and create the simplest and most seamless developer experience possible. This mantra has also been applicable to how we think about search, from the beginning with Atlas Text Search, to the announcement of the next paradigm with Atlas Vector Search. We have continued to expand this vision with the introduction of Search Nodes, initially launching on AWS, and then expanding to both Google Cloud and Microsoft Azure. Today we’re excited to take the next step in that journey with the announcement of multi-region availability on all three major cloud providers.

Search Nodes: Isolation and scale

As a quick refresher, Search Nodes provide dedicated infrastructure for Atlas Search and Vector Search workloads, enabling even greater control over search workloads. They also allow you to isolate and optimize compute resources to scale search and database needs independently, delivering better performance at scale and higher availability.

Since our announcements, we’ve been thrilled with the excitement around Search Nodes and the desire for better control, flexibility, and availability for scaling both Atlas Search and Vector Search workloads. Incorporating Search Nodes into your deployment delivers workload isolation, and the ability to optimize resource usage.

A visual of the evolution from the previous coupled architecture to dedicated nodes is shown below:

Graphic titled Better aligning workload size with resourcing. On the left of the graphic, is a cylinder labled coupled architecture, which contains search, database, and vector search. Arrows point from this left component over to the right, where there is a box titled dedicated search nodes. This side contains search, database, and scalability.
Figure 1: Improved workload sizing alignment and enhanced scalability with Search Nodes

Introducing Global Availability

Another tenet of our builder's journey is making sure the flexibility, scalability, and performance with Search Nodes are available to everyone, regardless of the cloud you’re using or cloud region. Today, we’re excited to officially announce multi-region availability for Search Nodes to allow anyone to better optimize resource usage regardless of location.

Now, with multi-region availability, you can take full advantage of global scalability by no longer being limited to one geographic area. Furthermore, you now have the peace of mind by having the redundancy needed to protect yourself in the case of any unforeseen outage event, whether due to technical issues or natural disasters that could cause data center downtime.

This image is a basic world map that has MongoDB leaf logos indicating where MongoDB cloud servers are available. There are several logos in North America, one in Brazil, several in Europe, a couple in Africa, one in Asia, and one in Australia.
Figure 2: Multi-region availability on all three major cloud providers

Here is a quick video tutorial about how to enable Search Nodes, as well as take advantage of multi-region availability:


Brief tutorial on how to enable multi-region Search Nodes

With today’s announcements we’re excited to bring the power and control of dedicated Search Nodes to people using all clouds and regions across the globe. We’re excited to see the continued adoption and improved results from having greater ubiquity across your search implementations.

As always, reach out to us with any feedback, as we’d love to hear what you think!