We currently upsert one document at a time in the receive portion of chunk migration. We should be able to batch insert documents and fall back to one-at-a-time if there is an error. This should be a performance win, but needs to be tested.
- depends on
-
SERVER-35849 Remove dependency of the write commands on `sharding_runtime_d`
- Closed
- is duplicated by
-
SERVER-22196 Sharding migrateClone phase should use insert and not upsert
- Closed
- is related to
-
SERVER-8835 _secondaryThrottle should GLE after every N documents rather than after each
- Closed
- related to
-
SERVER-32885 Overlap chunk clone application on the donor with fetching documents from the recipient
- Closed
-
SERVER-37080 Implement tuneable batch size for chunk migrations
- Closed