How spark handle hdfs compressed file and how to choose hdfs compression codec(splittable or not splittable) How spark handle hdfs compressed file and how to choose hdfs compression codec(splittable or not splittable) hadoop hadoop