I am doing a cube operation with count aggregation on a spark dataframe that has close to 1 million rows. I am using 4 columns for doing this cube operation. I notice that t
