Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Fixed
-
None
-
None
Description
The below code is a bad performer at scale since it has to memcpy the whole list of segments for each item in the batch instead of of just once per batch.
This is true for both computeBytesInMerges and getSegmentDescriptors.
for (int i = 0; i < batch; i++) { ArrayList#remove(0) }