問題文
A retail sales pipeline runs on a cluster with 4 executors that each have 4 cores. How many tasks can run at the same time, and what does the Apache Spark tuning guidance say about the number of partitions to aim for?
選択肢
- 16 tasks can run concurrently, and the guidance suggests exactly one partition per slot so that no task waits for a free slot.
- The concurrency depends on the number of input files, not on the cores.
- 16 tasks can run concurrently, and the guidance suggests 2 to 3 tasks per core, so more partitions than slots is deliberate.
- 4 tasks can run concurrently because one task uses one executor at a time, and the guidance suggests exactly one partition per executor so that each executor gets a single continuous unit of work with no scheduling overhead between the units.