How can we distribute a dataset into different tables based on their features/attributes for a scalable solution and perform required operations on the tables in that databa