On memory usage grow during sort spill

[yjshen]

Through examining the current implementation of multi-column sort's spill-to-disk strategies, I find we are asking for more memory during spill, which I think is worth discussing:

During the spill, Rows are created for comparison efficiency for each in-memory RecordBatch. Considering why we spill in the first place, does this Rows optimization increase the possibility of execution failure due to memory shortage?

Please share your insights or corrections if I misunderstand any part of the code.

[Omega359]

The comments on this issue may possibly be relevant (not 100% sure though)