Hi, I'm trying to figure out how the Spark standalone cluster mode works. Specifically, I'm looking at the code to see how the user's application jar makes it from the submission node, when it is on the local file system, to the driver and the executors.
>From what I can see, spark-submit causes the master to spin up a driver somewhere, and then the driver spins up executors. I couldn't figure out how the application jar goes through though. Could someone explain how this mechanism works? Thanks!