Of the following, which is the coarsest level in the Spark execution hierarchy?

Question

Options

• ASlot
• BJob
• CTask
• DStage
• EExecutor

• A
  2% (1)
• B
  93% (39)
• C
  2% (1)
• D
  2% (1)

Explanation

Job (B) sits at the top of Spark's execution hierarchy - it is spawned whenever an action (e.g., collect(), count()) is called on an RDD or DataFrame, and it encompasses all the work needed to produce that result.

• Stage (D) is one level below a Job; Spark splits each Job into Stages at shuffle boundaries, so a single Job may contain multiple Stages.
• Task (C) is the finest unit of execution - one Task processes one data partition and runs inside one Slot on an Executor.
• Executor (E) is a JVM process on a worker node - it's a resource concept, not a level in the logical execution hierarchy.
• Slot (A) is a single thread within an Executor, the smallest resource unit; Tasks are assigned to Slots at runtime.

Memory tip: Use the top-down acronym J → S → T - "Jobs Span Tasks" - to remember the execution hierarchy: one Job breaks into Stages, and each Stage breaks into Tasks.

No community discussion yet for this question.