-
Type: Task
-
Resolution: Unresolved
-
Priority: Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
Cluster Scalability
If I'm reading this correctly, the current queue size is numCursors*2 (I suspect that the original intent was number of results * 2). Given that a fetcher can get thousands of results in one go, the queue depth seems too low, and can potentially make the fetcher block unnecessarily instead of dumping everything to the queue and fetch the next results immediately.
- is related to
-
SERVER-94639 Make all sharding constants configurable
- Open