Free Databricks Associate-Developer-Apache-Spark Exam Dumps Questions & Answers

Which of the following code blocks immediately removes the previously cached DataFrame transactionsDf from memory and disk?

• A.array_remove(transactionsDf, "*")
• B.transactionsDf.clearCache()
• C.transactionsDf.persist()
• D.del transactionsDf
• E.transactionsDf.unpersist()
  (Correct)

In which order should the code blocks shown below be run in order to return the number of records that are not empty in column value in the DataFrame resulting from an inner join of DataFrame transactionsDf and itemsDf on columns productId and itemId, respectively?
1. .filter(~isnull(col('value')))
2. .count()
3. transactionsDf.join(itemsDf, col("transactionsDf.productId")==col("itemsDf.itemId"))
4. transactionsDf.join(itemsDf, transactionsDf.productId==itemsDf.itemId, how='inner')
5. .filter(col('value').isnotnull())
6. .sum(col('value'))

• A.4, 6
• B.4, 1, 2
• C.3, 1, 2
• D.3, 1, 6
• E.3, 5, 2

Which of the following describes a difference between Spark's cluster and client execution modes?

• A.In cluster mode, a gateway machine hosts the driver, while it is co-located with the executor in client mode.
• B.In cluster mode, executor processes run on worker nodes, while they run on gateway nodes in client mode.
• C.In cluster mode, the driver resides on a worker node, while it resides on an edge node in client mode.
• D.In cluster mode, the cluster manager resides on a worker node, while it resides on an edge node in client mode.
• E.In cluster mode, the Spark driver is not co-located with the cluster manager, while it is co-located in client mode.

The code block displayed below contains an error. The code block should count the number of rows that have a predError of either 3 or 6. Find the error.
Code block:
transactionsDf.filter(col('predError').in([3, 6])).count()

• A.Instead of filter, the select method should be used.
• B.Instead of a list, the values need to be passed as single arguments to the in operator.
• C.Numbers 3 and 6 need to be passed as string variables.
• D.The method used on column predError is incorrect.
• E.The number of rows cannot be determined with the count() operator.

Which of the following code blocks can be used to save DataFrame transactionsDf to memory only, recalculating partitions that do not fit in memory when they are needed?

• A.transactionsDf.clear_persist()
• B.transactionsDf.storage_level('MEMORY_ONLY')
• C.transactionsDf.cache()
• D.transactionsDf.persist()
• E.from pyspark import StorageLevel
  transactionsDf.cache(StorageLevel.MEMORY_ONLY)
• F.from pyspark import StorageLevel
  transactionsDf.persist(StorageLevel.MEMORY_ONLY)