-
Type: Improvement
-
Resolution: Unresolved
-
Priority: Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
Cluster Scalability
Currently, resharding cloning is implemented in a way that it does fetch batch and then insert batch sequentially in a loop. This means that no inserts are happening while fetching and vice versa. And if either of them blocks, the cloner is basically idle.
- is related to
-
SERVER-72144 Reshard cloner inserts gets slower as output collection gets larger
- Backlog