Hi!
Thanks for the library!
Could you write up some documentation on how those options work for ParallelStream.map?
My intuition says that as long as num_workers: 1, it's effectively not parallel, and worker_work_ratio is for managing the batch size passed to each worker for fine tuning acceptable parallel overhead?
Hi!
Thanks for the library!
Could you write up some documentation on how those options work for
ParallelStream.map?My intuition says that as long as
num_workers: 1, it's effectively not parallel, andworker_work_ratiois for managing the batch size passed to each worker for fine tuning acceptable parallel overhead?