Hi - I can see here Clustered Collections - Database Manual - MongoDB Docs:
Warning
Randomly generated key values may decrease a clustered collection’s performance.
Could you confirm if it is both read and write performance at risk or write performance only?
In my use case, I could benefit from increased read performance due to single-read architecture for “clustered collection” but me _id’s (keys) have to be random, and I do not care about write performance degradation.