Which of the following statements are true for Spark ?

		Spark RDD is fault-tolerant without the need for datareplication.
		wide transformations such as groupByKey() does not require datashuffling across Spark workers.
		A Spark application may trigger multiple jobs.
		Spark can not persist data in memory.

1. Which of the following statements are True ?

   		Unlike Spark, MapReduce is very suitable for iterativealgorithms.
   		In MapReduce framework, map tasks must complete before reducetask can start executing.
   		In Spark, a lineage graph plays a major role in achievingfault-tolerance for RDDs.
   		Pre-copy VM migration has lower downtime and higher migrationtime than post-copy VM migration.

Answer rating: 100% (QA)

Answer A The correct statements are wide transformations such as grou... View the full answer