I am curious about the tradeoffs that drove design of the partition/sort/shuffle (Elephant book p 208). Doubtless this has been tuned and measured and retuned, but I'd like to know what observations came about during the iterative optimization process to drive the final design. For example:
* Why does the mapper output create a single ordered file containing all partitions, as opposed to a file per group of partitions (which would seem to lend itself better to multi-core scaling), or even a file per partition?
* Why does the max number of streams to merge at once (is.sort.factor) default to 10? Is this obsolete? In my experience, so long as you have memory to buffer each input at 1MB or so, the merger is more efficient as a single phase.
* Why does the mapper do a final merge of the spill files do disk, instead of having the auxiliary process (in YARN) merge and stream data on the fly?
* Why do mappers sort the tuples, as opposed to only partitioning them and letting the reducers do the sorting?
Sorry if this is overly academic, but I'm sure a lot of people put a lot of time into the tuning effort, and I hope they left a record of their efforts.