Full-Text Search and Auto Scaling Coming To MongoDB Atlas

This is just one of the announcements made today as part of our CTO's keynote. Watch Eliot Horowitz's keynote to find out how it fits with our other announcements and vision for a data platform.

Among the many other announcements in the MongoDB World 2019 keynote were two MongoDB Atlas announcements which extend the power and reach of the managed platform for MongoDB users of any size. Full-Text Search enables users to leverage the power of Apache Lucene for text search. Auto Scaling builds upon current scaling options to enable an intelligent best-fit for workloads.

MongoDB Atlas Full-Text Search

MongoDB Atlas's new Full-Text Search service is set to bring the power of Lucene 8 to MongoDB. The new service is designed to end the reign of complex, application level integrations with external search engines. Instead, using MongoDB Atlas Full-Text Search, users and developers will get the productivity benefits of a transparent, full integrated search capability.


See Full-Text Search demonstrated at MongoDB World 2019 Keynote

MongoDB Atlas Full-Text Search can be used without managing the exporting of your data or configuring a separate search platform. Just activate the Full-Text Search option to your MongoDB Atlas cluster, and create your indexes. By default, it indexes all string fields in your collection but can be focussed down to index only particular fields. The configuration options for search include the ability to set up one or more analyzers for each index.

Queries for Full-Text Search leverage the MongoDB Query Language, allowing them to be used as with other MongoDB operators in powerful data processing pipelines. The $searchBeta aggregation stage allows for detailed control of the search process too.

The new Full-Text Search option requires MongoDB 4.2 or later and an M30 or above cluster and is currently in open beta. It will support replica sets and sharded clusters and is set to become generally available in the coming months.

If you want to check it out now, use the code MONGODB4DOT2 to get $200 of Atlas credit, spin up an M30 and discover for yourself the power and convenience of MongoDB Atlas Full Text Search. You'll also find more information in the documentation.

MongoDB Atlas Auto Scaling

Auto Scaling is a new feature coming to Atlas which reduces the effort to get your database scaling to the right size for your applications. MongoDB Atlas has always made it easy to scale through the API or UI with the service and automated that process to ensure that it had no impact on applications. All that an administrator had to do was review the workload, the performance metrics and any alerts and then decide on how to respond to those requirements.

Previously, Atlas had offered auto-scaling for disk storage. With the new Auto Scaling option, that capability expands to handle CPU and RAM requirements. With Auto Scaling enabled, that tracking of key metrics is handled by Atlas in real time. These metrics feed a process that is a combination of predictive modeling and the proven practices that come from managing tens of thousands of MongoDB deployments.

The results of this process turn into automated scaling up or down actions being applied to the database being managed. With this in place, users and administrators can rest assured that their clusters will be adequately resourced for their workloads. Through the Auto Scaling UI, administrators will be able to set limits and alerts to how far the auto scaling process can go, ensuring they retain control.

Auto Scaling is currently in private beta and is expected to be widely available later this summer.