Explain the Apache Spark™ Architecture Execution Hierarchy.. ASSOCIATE-DEVELOPER-APACHE-SPARK Practice Questions (Page 2)

Questions 6–10

1. 6application · easy

   A Spark application reads a JSON file, applies a filter, then a groupBy, then a join with another DataFrame, and finally calls show(). How many jobs will be triggered by this application?

   • AFour jobs, one for each operation in the pipeline.
   • BTwo jobs, one for the read and one for the show().
   • CThree jobs, one for each transformation.
   • DOne job, because show() is the only action.
   Reveal Solution Discussion Select an answer first
2. 7application · medium

   A developer has written a Spark application that reads a large CSV file, applies several transformations (filter, map, and select), and then writes the result to a Delta table. The developer notices that the Spark UI shows no jobs have been executed even though the transformations appear to have run. What is the reason for this?

   • AThe transformations are executed on the driver, not on the executors, so no jobs are created.
   • BThe transformations are lazy and no job is triggered until an action like write is called.
   • CThe write operation is also lazy, so the application must call an additional action like count() to trigger execution.
   • DThe read operation is also lazy, so the entire application is waiting for the driver to collect the data.
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A developer is debugging a Spark application that reads from a Kafka stream and writes to a Delta table. The application uses a streaming query with a trigger that processes data every 5 seconds. The developer notices that the Spark UI shows a new job every 5 seconds. What is the reason for this?

   • AThe write to Delta is a transformation, so it creates a new job for each record.
   • BThe trigger is set to 5 seconds, so Spark automatically creates a new job every 5 seconds even if there is no data.
   • CThe Kafka read is an action that triggers a job for each partition.
   • DEach micro-batch is an action that triggers a new job.
   Reveal Solution Discussion Select an answer first
5. 9expert · medium

   A Spark job is running with 4 stages. The first stage has 100 tasks, the second has 200 tasks, the third has 100 tasks, and the fourth has 50 tasks. The cluster has 50 cores. What is the maximum concurrency during the second stage?

   • A50 tasks, but only if the stage is the only one running.
   • B100 tasks
   • C200 tasks
   • D50 tasks
   Reveal Solution Discussion Select an answer first
6. 10foundation · easy

   Which component is responsible for converting a logical DAG of stages into tasks and submitting them to executors?

   • ACluster Manager
   • BTask Scheduler
   • CDriver
   • DDAG Scheduler
   Reveal Solution Discussion Select an answer first