Online Archive

Online Archive is a feature in MongoDB Atlas that allows you to automatically archive infrequently accessed data from your live Atlas clusters to fully managed cloud object storage. This helps optimize storage costs and performance without sacrificing data accessibility. You can set custom rules to define when and how data is archived and seamlessly query live and archived data from a single endpoint.

Archiving data begins with setting up an archiving rule by specifying a namespace (database and collection) and choosing one of the following options:

• Date match: Define a date field within the documents and set an age limit (number of days past the date field) when archival should begin
• Custom filter: Write a custom query to select documents for archiving.

Once the rule is established, Atlas automatically archives documents matching the criteria for fully managed cloud object storage. You can then query your live cluster and archived data using a unified endpoint.

Online Archive is available on all dedicated Atlas clusters of M10 or higher tiers. Archived data is stored on MongoDB’s fully managed cloud object storage and is encrypted by MongoDB’s encryption key. To query an Online Archive, you can use a public connection or AWS PrivateLink.

Archival jobs run every five minutes. Documents that match the archiving rule are stored in a temporary collection on the Atlas cluster, then archived in files up to 100MB, with a total of up to 2GB per five-minute interval. This approach ensures that archival jobs don’t overwhelm the cluster’s resources.

No, Online Archive is not a replacement for backups—it is designed to support a data archiving strategy. To ensure complete data protection, Online Archive should be used alongside your backup strategy, ensuring that any data moved from your Atlas cluster to Online Archive has already been backed up.