Q39
Single choice
A Spark developer is developing a Spark application to monitor task performance across a cluster. One of the application's requirements is to track the maximum processing time for tasks on each worker node and consolidate this information on the driver for further analysis.
Which technique should the developer use to achieve this?