Add a thing to generate executor memory using input data size + container sizes #5
Description
Input data size could be a user hint or based off of previous run or hadoop file or Hive stastitics ala Spark's own cost-based-query-planner.
{{ message }}