I am trying to understand about task time and GC time in executor section in Spark Databricks. For all task it took around 20 mins actual time to complete execution of all tasks. But when I see it shows in UI as 4.7 hr. I want to understand how Gc time and task time is calculated here because in reality it completed in lot more lesser time than what displayed.
CodePudding user response:
Since Tasks run in parallel, the task time can be way higher than the actual time that passed.
As you can see there are 12 active tasks in this screenshot, running in parallel.