
Explanation:
Spark's design includes robust fault tolerance for worker nodes but does not automatically recover from driver failures. The incorrect statements about Spark's stability are those that misrepresent these capabilities. Option A is incorrect because Spark cannot support the loss of any set of worker nodes without limitation; it relies on having some workers available to rerun tasks. Option E is incorrect because Spark does not reassign the driver to a worker node if the driver's node fails; instead, the application fails. Options B, C, and D correctly describe Spark's behavior regarding task rerunning, data recomputation, and disk spilling, respectively.
Ultimate access to all questions.
Which of the following statements about Spark's reliability is false?
A
Spark is designed to support the loss of any set of worker nodes.
B
Spark will rerun any failed tasks due to failed worker nodes.
C
Spark will recompute data cached on failed worker nodes.
D
Spark will spill data to disk if it does not fit in memory.
E
Spark will reassign the driver to a worker node if the driver’s node fails.
No comments yet.