Big Data Practice Quiz: Spark Quiz 1

1. In Spark, each job gets divided into smaller sets of tasks that depend on each other and are called which of the following?

• containers
• units
• ladders
• stages

2. What is the maximum recommended allocation of memory for Spark (leaving the remainder for the operating system and buffer cache)?

• 50 percent
• 60 percent
• 75 percent
• 90 percent

3. Which organization are Apache Spark, Spark, and the Spark logo trademarks of?

• Apache Software Foundation
• GNU Project
• Free Software Foundation
• Free and Open Source Software

4. In Spark, what are RDD operations that produce non-RDD values known as?

• Blossoms
• Actions
• Decrees
• Compiles

5. Which of the following does the Spark driver program use to interact with the worker nodes?

• Executor
• System Governor
• Cluster Manager
• Cluster Superintendent
• Proprietor

6. Spark scales to tens of CPU cores per machine because it performs minimal sharing between threads. For optimization purposes, how many cores per machine should you provision?

• 12-20
• 10-12
• 4-8
• 20-30
• 8-16

7. Which of the following Spark metric instances represent the ApplicationMaster when running on YARN?

• application
• applicationSnitch
• master
• applicationMaster

8. When RPC authentication is turned on via spark.authenticate, where is the secret key stored?

• spark.authenticate.passwd
• spark.authenticate.secret
• spark.authenticate.shadow
• spark.authenticate.shared

9. By default, what is the default port to which the web interface of the Spark history server binds?

• 4403
• 4043
• 18080
• 43

10. Under cluster mode, each application gets its own which kind of process?

• Overseer
• Boss
• Executor
• Chief