For a spark dstream application I need to access previous batch dataframes/rdds , So i have tried to recreate a toy example for better understanding.
Usecase
