I am currently using Apache Spark with Graphx, I have noticed lately that when I run my application with a lots of data the application is using a large part of my disk, for example before I start the program the disk is around 8 GB and during the application runs it goes down to 1 GB, when I close the application the disk is restored but not in full. I have lost some GB, at first I though that it had to do with swap memory and logs, but I can not find what is stored to my disk after the execution of the application.
Can someone explain why is this happening?
