What is the basic difference between partitioning datasets by key or
grouping them by key ?

Does it make a difference in terms of paralellism ?

Thx

Reply via email to