In-Use Encryption is best applied selectively to those fields of your documents that you classify as containing the most sensitive data, such as PII or PHI.

Using Client-Side FLE or Queryable Encryption alongside in-transit and at-rest encryption provides encryption of data throughout its lifecycle, using complementary approaches that provide a defense-in-depth security posture to address different threat models.

• In-transit encryption protects all data traversing the network but does not encrypt data in-use or at-rest.
• At-rest encryption protects all stored data but does not encrypt data in-use or in-transit.
• With in-use encryption, your most sensitive data never leaves your application in plaintext. Fields that are encrypted client-side cannot be decrypted by the server and remain encrypted in-transit, at-rest, and in-use even as queries are being processed.

Your encryption keys can be stored in any cloud provider of your choosing and do not need to be in the same cloud provider as your data. For example, you can store your data in Azure but have your encryption keys in AWS KMS, or have your data in AWS but your keys in GCP Cloud KMS. This approach applies both to at-rest encryption and in-use encryption capabilities.

MongoDB Enterprise supports KMIP-enabled key providers for encryption at-rest. Cloud-provided KMS (Key Management Systems) is not supported.

Queryable Encryption uses a fast encrypted search algorithm to add additional encrypted data structures on the server side, enabling the processing of expressive queries on encrypted data. Since the database does the query processing, there’s no need to bring extra results back to the client or write additional application code for client-side query handling.

Client-Side Field Level Encryption provides related functionality, in that it encrypts data on the client side before inserting it into the database. Querying is limited to equality queries and deterministic encryption must be used for equality.

The Queryable Encryption supports equality queries on encrypted data as of the 7.0 release.

Future releases will support range, prefix, suffix, and substring query types.