Each stage of a Spark task can be divided into jobs, and the division mark is shuffle.

Question

Options

• ATrue
• BFalse

• A
  18% (4)
• B
  82% (18)

Explanation

Option B is correct because the statement has the Spark execution hierarchy inverted. In Spark, it is a Job that gets divided into Stages - not the other way around. The shuffle operation (triggered by wide transformations like groupByKey or reduceByKey) marks the boundary between stages within a job, not between jobs within a stage.

Option A is wrong because stages are smaller units within a job; a stage cannot contain or be divided into jobs, which are higher-level units triggered by actions like collect() or save().

Memory tip: Think of the hierarchy top-down - Jobs → Stages → Tasks (just remember JST: "Just Shuffle There"). Shuffles split stages, actions split jobs, and partitions split tasks. The question reverses the first two levels, which is the trap.

No community discussion yet for this question.