I need to know if there is any way to fetch n random records from first partition in spark. For example if i want to fetch 50 records and there are total 10000 records in datase