We have a collection on production environment and it is having only 126 records, but it is taking maximum 6 seconds for a single findOne operation. 

We are using mongodb version 3.4 , and it is having very slow response time for even smaller find operations. All the find operations are using IXSCAN with proper index utilisation, and all read and write operations are taking more than 1 second even with smaller document size having 10-12 fields.

We have 37GB of data and our production environment having 3 member replica set with primary, secondary, secondary configuration.

Please update us with possible optimisation methods.