As part of PM-3014 (and its backports to 5.0/6.0 done under PM-3001) we changed the amount of memory used during the cloning phase on the recipient shard, going from a constant amount to a variable one that depends on the speed of fetching data (producer) and locally inserting it (consumer).
Cloning a chunk of 80GB we saw spikes of memory of 52GB.
- is related to
-
SERVER-80977 Abort chunk migration after cloning more data than expected
- Backlog
-
SERVER-67183 Higher throughput Chunk Migration - reimplement for fetching and insertion
- Closed