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
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