The cloud search engine is a mongodb compatible database intended to run as a standalone database or as a companion database to your mongodb instance.
When used as a companion database, the cloud search engine will be used mainly for search and sql analytics. For this use case, the cloud search engine can replicate your data in real-time from your mongodb database.
When using in standalone mode, it is important to understand the cloud search engine does not support map reduce or in database javascript expressions. The cloud search engine is SQL centric. Also the error messages and codes that will be returned from the cloud search engine for some cases will be different than what is produced by a mongodb instance. It is important to test your app thoroughly to ensure the compatibility does not break anything.
After creating a cloud search engine from the dashboard, you can retrieve the connection string and set it as an environment variable for your app. You can then connect to the server using a mongo client.
The documentation on data synchronization will guide you on how to replicate your data from the mongodb database to the cloud search engine in real-time.
It is important to note, although the cloud search engine replicates your data from the mongodb database, your mongodb instance actually has no knowledge of the cloud search engine as a replica in its set.
After creating a cloud search engine from the dashboard, you can retrieve the connection string and then set the MONGO_URI, DATABASE_URI environment variable for your app. This allows your app to use the cloud search engine as its primary database.
The compatiblity sheet details the features from mongodb currently supported on the cloud search engine.