Ministère de l’Enseignement Supérieur et de la Recherche Scientifique

Université de Carthage
Institut Supérieur des Technologies de l’Information et de la Communication

Classes : LFSI-3 Nombre de pages : 8

Devoir Surveillé
Matière : BIG DATA
Enseignants : Zayneb Trabelsi Date : 09/03/2020 Durée : 1h30
Documents : autorisés 
Nom : ………………………………………. Groupe : …………………………… : non autorisés 

Prénom : …………………………………..

Please answer the following multiple-choice questions:

1. Under MapReduce V1, what is one purpose of the TaskTracker?

A1. ☐ Coordinates MapReduce jobs

A2. ☒ Manages storage and transmission of intermediate output

A3. ☐ Accepts MapReduce jobs submitted by clients

2. What is the default number of replicas in a Hadoop system?

A1. ☐ two

A2. ☒ three

A3. ☐ four

3. Spark provides the shell in two programming languages: (select two.)

A1. ☒ Scala

A2. ☒ Python

A3. ☐ R

A4. ☐ SQL

4. One of the driving principal of Hadoop is that the data is brought to the program?

A1. ☐ True

A2. ☒ False

1
 5. Which database is a NoSQL key-value storage database?

A1. ☒ REDIS

A2. ☐ HBase

A3. ☐ Neo4j

6. Which of the following options is NOT CORRECT ?

A1. ☐ Big data solutions are ideal for analyzing not only raw structured data, but semi structured and

unstructured data from a wide variety of sources.

A2. ☒ Big data solutions are ideal for Online Transaction Processing (OLTP) environments.

A3. ☐ Big data solutions are ideal for iterative and exploratory analysis when business measures on
data are not predetermined.

7. True or False : CREATE statement in HBase requires the name of the table and the list of its
columns

a. ☐ True

b. ☒ False

8. What is NOT true regarding Hive External Tables?

a. ☐ data is stored outside the Hive warehouse directory

b. ☒ data is stored outside the Hadoop cluster

c. ☐ when dropping table, only the metadata is deleted, and data is left untouched

d. ☐ All of the above.

9. If Active NameNode fails, then which of the following Node takes all the responsibility of
active node?

A1. ☐ Secondary NameNode

A2. ☒ Standby NameNode

A3. ☐ All of the above.

10. Which of the following statement is true about metadata?

A1. ☐ Metadata shows the structure of HDFS directories/files

A2. ☐ FsImage & EditLogs are metadata files

A3. ☐ Metadata contain information like number of blocks, their location, replicas

A4. ☒ All of the above

2
 11. The source of HDFS architecture in Hadoop originated as

A1. ☒ Google distributed filesystem

A2. ☐ Yahoo distributed filesystem

A3. ☐ Facebook distributed filesystem

A4. ☐ None of the above

12. What does Big Data represent?

A1. ☐ A Hadoop feature capable of processing vast amounts of data in-parallel on large clusters of
commodity hardware in a reliable, fault-tolerant manner.

A2. ☒ A concept and platform of technologies with the characteristics of the 5 Vs that is able to handle
large amounts of unstructured, semi-structured, and structured raw data unlike traditional systems.

A3. ☐ A database feature capable of converting pre-existing structured data into unstructured raw
data.

A4. ☐ Only data stored in the BIGDATA table in any relational database.

13. Which is not a principle of Hadoop

A1. ☐ Bring processing to Data

A2. ☐ Replicate data blocks to multiple nodes

A3. ☒ None of the above

14. Which is the 5th V that is the real purpose of working with Big Data to obtain business
insight?

A1. ☐ Volume

A2. ☐ Variety

A3. ☒ Value

A4. ☐ None of the above

15. What is Hadoop ?

A1. ☒ An open-source software framework for distributed storage and distributed processing of Big
Data on clusters of commodity hardware.

A2. ☐ It was conceived for high-end, expensive hardware.

A3. ☐ An environment for on-line transaction processing.

A4. ☐ It consists of 3 sub projects: MapReduce, Hive and Hadoop Common.

A5. ☐ All of the above

3
 16. What are the main components of Hadoop ? (Select four.)

A1. ☒ Map Reduce

A2. ☐ HBase

A3. ☒ Hadoop Distributed File System

A4. ☒ YARN

A5. ☒ Hadoop Common

17. What is NOT true about Hadoop Distributed File System (HDFS) ?

A1. ☐ Can create, delete, copy but not update

A2. ☐ Files split into blocks

A3. ☐ Data access through MapReduce

A4. ☒ Designed for random access not streaming reads

18. What is NOT true about the Namenode StartUp ?

A1. ☐ NameNode reads fsimage in memory

A2. ☐ NameNode applies editlog changes

A3. ☐ NameNode exits safemode when 99.9% of blocks have at least one copy accounted for.

A4. ☒ NameNode stores data blocks

A5. ☐ NameNode waits for block data from data nodes

19. What is an optional MapReduce Task ?

A1. ☐ Map

A2. ☐ Shuffle

A3. ☐ Reduce

A4. ☒ Combiner

20. Which is the Hadoop-related Apache project that utilizes an in-memory architecture to run
applications faster than MapReduce?
A1. ☐ Hive

A2. ☐ Python

A3. ☒ Spark

A4. ☐ Pig

4
 21. Apache Spark can run on which two of the following cluster managers? (Select two.)

A1. ☒ Apache Mesos

A2. ☒ Hadopp Yarn

A3. ☐ Linux cluster manager

A4. ☐ onesies

22. What Hadoop command is used to copy files from HDFS to local file system:

a. ☐ hadoop fs -put

b. ☒ hadoop fs -get

c. ☐ hadoop copyfromlocal

23. What is NOT true about HDFS-2 Namenode HA

A1. ☐ Memory state of Standby Namenode is very close to Active Namenode

A2. ☒ Datanodes send heartbeats to the Active Namenode only

A3. ☐ Unlike the Secondary NameNode, the Standby NameNode allows a fast failover to a new
NameNode in the case that a machine crashes

A4. ☐ None of the above

24. True or False: Hive comes with an HBase storage handler.

A1. ☒ True

A2. ☐ False

25. The clause that indicates the storage file/record format on HDFS.

A1. ☒ stored as

A2. ☐ stored in

A3. ☐ stored to

A4. ☐ None of the above

26. Which two factors in a Hadoop cluster increase performance most significantly? (Select
two.)

A1. ☐ immediate failover of failed disks

A2. ☐ large number of small data files

A3. ☒ high-speed networking between nodes

A4. ☐ data redundancy on management nodes

A5. ☒ parallel reading of large data files

5
 27. What is not true about HBase?

A1. ☐ An industry leading implementation of Google's Big Table design

A2. ☐ An open source Apache Top Level Project

A3. ☒ A RDBMS that powers some of the leading sites on the Web

A4. ☐ A NoSQL data store

A5. ☐ None of the above

28. Under the HDFS architecture, what is one purpose of the NameNode?

A1. ☐ to coordinate MapReduce jobs

A2. ☐ to periodically report status to DataNode

A3. ☒ to regulate client access to files

29. What is the "scan" command used for in HBase?

a. ☐ to get detailed information about the table
b. ☐ to report any inconsistencies in the database
c. ☐ to list all tables in Hbase
d. ☒ to view data in an Hbase table

30. Hadoop environments are optimized for:

A1. ☐ Processing transactions (random access).

A2. ☐ Low latency data access.

A3. ☒ Batch processing on large files.

A4. ☐ Intensive calculation with little data.

31. How are data stored in a Hadoop cluster?

A1. ☒ The data is divided into blocks, and copies of these blocks are replicated across multiple servers
in the Hadoop cluster.

A2. ☐ The data converted into a single block, and the block is stored in just one of the servers in the
Hadoop cluster.

A3. ☐ The data is divided into blocks, each block is stored in a different server in the Hadoop cluster,
but the blocks are not replicated.

A4. ☐ The data is converted into a single block, and copies of this block are replicated across multiple
servers in the Hadoop cluster.

6
 32. Which of the following options best describes the proper usage of MapReduce jobs in
Hadoop environments?

A1. ☒ MapReduce jobs are used to process vast amounts of data in-parallel on large clusters of
commodity hardware in a reliable, fault-tolerant manner.

A2. ☐ MapReduce jobs are used to process small amounts of data in-parallel on expensive hardware,
without fault-tolerance.

A3. ☐ MapReduce jobs are used to process structured data in sequence, with fault-tolerance.

33. In a traditional Hadoop stack, which of the following components provides data warehouse
infrastructure and allows SQL developers and business analysts to leverage their existing SQL
skills?

A1. ☐ MapReduce.

A2. ☒ Hive.

A3. ☐ Zookeeper.

34. Following the most common HDFS replica placement policy, when the replication factor is
three, how many replicas will be located on the local rack?

A1. ☐ two

A2. ☒ one

A3. ☐ three

35. Which primary computing bottleneck of modern computers is addressed by Hadoop?

A1. ☐ MIPS

A2. ☒ disk latency

A3. ☐ limited disk capacity

36. What are two primary limitations of MapReduce v1? (Select two.)
A1. ☒ Resource utilization

A2. ☒ Scalability

A3. ☐ Workloads limited to MapReduce

A4. ☐ Number of TaskTrackers limited to 1000

A5. ☐ TaskTrackers can be a bottleneck to Mapreduce jobs

7
 37. In Spark, to do an RDD transformation by reading in a file that was previously loaded to
HDFS, we type the following:
val pp = sc.textFile("Gutenberg/Pride_and_Prejudice.txt")

True or False: “pp” is just a pointer to the RDD

a. ☒ True

b. ☐ False

38. Which Hbase command lists individual row data?

a. ☐ scan

b. ☐ show all

c. ☒ get

d. ☐ All of the above.

39. Which command must be sometimes executed before deleting an HBase table or changing
its settings ?

a. ☐ alter

b. ☒ disable

c. ☐ None of the above

40. The following command is executed in HDFS.

hadoop fs -copyFromLocal /home/student/labfiles/*.txt Gutenberg

This command can also be executed as:

a. ☒ hadoop fs -put /home/student/labfiles/*.txt Gutenberg

b. ☐ hadoop fs -get /home/student/labfiles/*.txt Gutenberg

c. ☐ None of the above