Question: 1

Where must a Spark configuration be set up first?

Your answer

A Notebook

B Db2 Warehouse

C IBM Cloud

D Watson Data Platform

Question: 2
When sharing a notebook, what will always point to
the most recent version of the notebook?
Your answer

A Watson Studio homepage

B The permalink

C The Spark service

D PixieDust visualization

Question: 3
When creating a Watson Studio project, what do you
need to specify?
Your answer

A Spark service

B Data service

C Collaborators
Your answer

D Data assets

Question: 4
You can import preinstalled libraries if you are using
which languages? (Select two.)
(Please select ALL that apply)
Your answer

A R

B Python

C Bash

D Rexx

E Scala

Question: 5
Who can control a Watson Studio project assets?
Your answer

A Viewers

B Editors

C Collaborators

D Tenants

Question: 6
Which environmental variable needs to be set to
properly start ZooKeeper?
Your answer

A ZOOKEEPER_APP

B ZOOKEEPER_DATA

C ZOOKEEPER

D ZOOKEEPER_HOME

Question: 7
Which is the primary advantage of using column-
based data formats over record-based formats?
Your answer

A better compression using GZip

B supports in-memory processing

C facilitates SQL-based queries

D faster query execution

Question: 8
What is the primary purpose of Apache NiFi?
Your answer

A Collect and send data into a stream.

B Finding data across the cluster.

C Connect remote data sources via WiFi.

D Identifying non-compliant data access.

Question: 9
What are three examples of Big Data? (Choose
three.)
(Please select ALL that apply)
Your answer

A cash register receipts

B web server logs

C inventory database records

D bank records

E photos posted on Instragram

F messages tweeted on Twitter

Question: 10
What ZK CLI command is used to list all the ZNodes
at the top level of the ZooKeeper hierarchy, in the
ZooKeeper command-line interface?
Your answer

A get /

B create /

C listquota /

D ls /

Question: 11
What is the default data format Sqoop parses to
export data to a database?
Your answer

A JSON

B CSV

C XML

D SQL

Question: 12
Under the MapReduce v1 architecture, which
function is performed by the TaskTracker?
Your answer

A Keeps the tasks physically close to the data.

B Pushes map and reduce tasks out to DataNodes.

C Manages storage and transmission of intermediate output.

D Accepts MapReduce jobs submitted by clients.

Question: 13
Which statement describes "Big Data" as it is used
in the modern business world?
Your answer

A Indexed databases containing very large volumes of historical data used for comp

B Non-conventional methods used by businesses and organizations to capture, ma

C Structured data stores containing very large data sets such as video and audio st

D The summarization of large indexed data stores to provide information about pote
Question: 14
Under the MapReduce v1 architecture, which
function is performed by the JobTracker?
Your answer

A Runs map and reduce tasks.

B Accepts MapReduce jobs submitted by clients.

C Manages storage and transmission of intermediate output.

D Reports status to MasterNode.

Question: 15
Which statement is true about the Hadoop
Distributed File System (HDFS)?
Your answer

A HDFS is a software framework to support computing on large clusters of compute

B HDFS provides a web-based tool for managing Hadoop clusters.

C HDFS links the disks on multiple nodes into one large file system.

D HDFS is the framework for job scheduling and cluster resource management.

Question: 16
How does MapReduce use ZooKeeper?
Your answer

A Coordination between servers.

B Aid in the high availability of Resource Manager.

C Master server election and discovery.

Your answer

D Server lease management of nodes.

Question: 17
Which two Spark libraries provide a native shell?
(Choose two.)
(Please select ALL that apply)
Your answer

A Python

B Scala

C C#

D Java

E C++

Question: 18
What is an authentication mechanism in
Hortonworks Data Platform?
Your answer

A IP address

B Preshared keys

C Kerberos

D Hardware token

Question: 19
What is Hortonworks DataPlane Services (DPS) used
for?
Your answer

A Manage, secure, and govern data stored across all storage environments.

B Transform data from CSV format into native HDFS data.

C Perform backup and recovery of data in the Hadoop ecosystem.

D Keep data up to date by periodically refreshing stale data.

Question: 20
What must be done before using Sqoop to import
from a relational database?
Your answer

A Copy any appropriate JDBC driver JAR to $SQOOP_HOME/lib.

B Complete the installation of Apache Accumulo.

C Create a Java class to support the data import.

D Create an empty database for Sqoop to access.

Question: 21
What is the native programming language for Spark?
Your answer

A Scala

B C++

C Java

D Python

Question: 22
Which Hortonworks Data Platform (HDP) component
provides a common web user interface for
applications running on a Hadoop cluster?
Your answer

A YARN

B HDFS

C Ambari

D MapReduce

Question: 23
Which Spark RDD operation returns values after
performing the evaluations?
Your answer

A Transformations

B Actions

C Caching

D Evaluations

Question: 24
Which two are use cases for deploying ZooKeeper?
(Choose two.)
(Please select ALL that apply)
Your answer

A Configuration bootstrapping for new nodes.

B Managing the hardware of cluster nodes.

Your answer

C Storing local temporary data files.

D Simple data registry between nodes.

Question: 25
In a Hadoop cluster, which two are the result of
adding more nodes to the cluster? (Choose two.)
(Please select ALL that apply)
Your answer

A DataNodes increase capacity while NameNodes increase processing power.

B It adds capacity to the file system.

C Scalability increases by a factor of x^N-1.

D Capacity increases while fault tolerance decreases.

E It increases available processing power.

Question: 26
Which Spark RDD operation creates a directed
acyclic graph through lazy evaluations?
Your answer

A Distribution

B GraphX

C Transformations

D Actions

Question: 27
Which feature allows application developers to
easily use the Ambari interface to integrate Hadoop
provisioning, management, and monitoring
capabilities into their own applications?
Your answer

A REST APIs

B Postgres RDBMS

C Ambari Alert Framework

D AMS APIs

Question: 28
What is one disadvantage to using CSV formatted
data in a Hadoop data store?
Your answer

A Columns of data must be separated by a delimiter.

B Fields must be positioned at a fixed offset from the beginning of the record.

C It is difficult to represent complex data structures such as maps.

D Data must be extracted, cleansed, and loaded into the data warehouse.

Question: 29
Which element of Hadoop is responsible for
spreading data across the cluster?
Your answer

A YARN

B MapReduce
Your answer

C AMS

D HDFS

Question: 30
Which component of the Apache Ambari
architecture stores the cluster configurations?
Your answer

A Authorization Provider

B Ambari Metrics System

C Postgres RDBMS

D Ambari Alert Framework

Question: 31
Which two are examples of personally identifiable
information (PII)? (Choose two.)
(Please select ALL that apply)
Your answer

A Time of interaction

B Medical record number

C Email address

D IP address

Question: 32
Under the MapReduce v1 architecture, which
element of the system manages the map and reduce
functions?
Your answer

A SlaveNode

B JobTracker

C MasterNode

D StorageNode

E TaskTracker

Question: 33
Which component of the HDFS architecture
manages storage attached to the nodes?
Your answer

A NameNode

B StorageNode

C DataNode

D MasterNode

Question: 34
Which of the "Five V's" of Big Data describes the
real purpose of deriving business insight from Big
Data?
Your answer

A Volume
Your answer

B Value

C Variety

D Velocity

E Veracity

Question: 35
Which component of the Spark Unified Stack
supports learning algorithms such as, logistic
regression, naive Bayes classification, and SVM?
Your answer

A Spark Learning

B Spork

C Spark SQL

D MLlib

Question: 36
Which two descriptions are advantages of Hadoop?
(Choose two.)
(Please select ALL that apply)
Your answer

A able to use inexpensive commodity hardware

B intensive calculations on small amounts of data

C processing a large number of small files

Your answer

D processing random access transactions

E processing large volumes of data with high throughput

Question: 37
Which two of the following are row-based data
encoding formats? (Choose two.)
(Please select ALL that apply)
Your answer

A CSV

B Avro

C ETL

D Parquet

E RC and ORC

Question: 38
Which statement describes the action performed by
HDFS when data is written to the Hadoop cluster?
Your answer

A The data is spread out and replicated across the cluster.

B The data is replicated to at least 5 different computers.

C The MasterNodes write the data to disk.

D The FsImage is updated with the new data map.

Question: 39
Under the MapReduce v1 architecture, which
element of MapReduce controls job execution on
multiple slaves?
Your answer

A MasterNode

B JobTracker

C SlaveNode

D TaskTracker

E StorageNode

Question: 40
Which component of the Spark Unified Stack
provides processing of data arriving at the system in
real-time?
Your answer

A MLlib

B Spark SQL

C Spark Streaming

D Spark Live

Question: 41
Which two registries are used for compiler and
runtime performance improvements in support of
the Big SQL environment? (Choose two)
(Please select ALL that apply)
Your answer

A DB2ATSENABLE

B DB2FODC

C DB2COMPOPT

D DB2RSHTIMEOUT

E DB2SORTAFTER_TQ

Question: 42
Which script is used to backup and restore the Big
SQL database?
Your answer

A bigsql_bar.py

B db2.sh

C bigsql.sh

D load.py

Question: 43
You need to create a table that is not managed by
the Big SQL database manager. Which keyword
would you use to create the table?
Your answer

A STRING

B BOOLEAN
Your answer

C SMALLINT

D EXTERNAL

Question: 44
Which two of the following data sources are
currently supported by Big SQL? (Choose two)
(Please select ALL that apply)
Your answer

A Oracle

B PostgreSQL

C Teradata

D MySQL

E MariaDB

Question: 45
Which port is the default for the Big SQL Scheduler
to get administrator commands?
Your answer

A 7055

B 7054

C 7052

D 7053

Question: 46
Which tool should you use to enable Kerberos
security?
Your answer

A Hortonworks

B Ambari

C Apache Ranger

D Hive

Question: 47
Which two options can be used to start and stop Big
SQL? (Choose two)
(Please select ALL that apply)
Your answer

A Scheduler

B DSM Console

C Command line

D Java SQL shell

E Ambari web interface

Question: 48
Which command is used to populate a Big SQL
table?
Your answer

A CREATE
Your answer

B QUERY

C SET

D LOAD

Question: 49
Which feature allows the bigsql user to securely
access data in Hadoop on behalf of another user?
Your answer

A Impersonation

B Privilege

C Rights

D Schema

Question: 50
Which command would you run to make a remote
table accessible using an alias?
Your answer

A SET AUTHORIZATION

B CREATE SERVER

C CREATE WRAPPER

D CREATE NICKNAME

Question: 51
The Big SQL head node has a set of processes
running. What is the name of the service ID running
these processes?
Your answer

A Db2

B hdfs

C user1

D bigsql

Question: 52
Which file format contains human-readable data
where the column values are separated by a
comma?
Your answer

A Parquet

B ORC

C Delimited

D Sequence

Question: 53
Which Big SQL authentication mode is designed to
provide strong authentication for client/server
applications by using secret-key cryptography?
Your answer

A Public key
Your answer

B Flat files

C Kerberos

D LDAP

Question: 54
Which type of foundation does Big SQL build on?
Your answer

A Jupyter

B Apache HIVE

C RStudio

D MapReduce

Question: 55
You need to monitor and manage data security
across a Hadoop platform. Which tool would you
use?
Your answer

A SSL

B HDFS

C Hive

D Apache Ranger

Question: 56
What can be used to surround a multi-line string in a
Python code cell by appearing before and after the multi-
line string?
Your answer

A """

B "

Question: 57
For what are interactive notebooks used by data
scientists?
Your answer

A Packaging data for public distribution on a website.

B Quick data exploration tasks that can be reproduced.

C Providing a chain of custody of all data.

D Bulk loading data into a database.

Question: 58
What Python statement is used to add a library to
the current code cell?
Your answer

A pull

B import

C load
Your answer

D using

Question: 59
What Python package has support for linear algebra,
optimization, mathematical integration, and
statistics?
Your answer

A NLTK

B Pandas

C NumPy

D SciPy

Question: 60
Which three main areas make up Data Science
according to Drew Conway? (Choose three.)
(Please select ALL that apply)
Your answer

A Traditional research

B Machine learning

C Substantive expertise

D Math and statistics knowledge

E Hacking skills
1- Select all the components of HDP which provides data access capabilities

 Pig
 Sqoop
 Flume
 MapReduce
 Hive

2- Select the components that provides the capability to move data from relational database
into Hadoop.

 Sql
 Sqoop
 Hive
 Kafka
 Flume

3- Managing Hadoop clusters can be accomplished using which component?

 Ambari
 HBase
 Phoenix
 Hive
 Sqoop

4- True or False: The following components are value-add from IBM: Big Replicate, Big SQL,
BigIntegrate, BigQuality, Big Match

 TRUE
 FALSE

5- True or False: Data Science capabilities can be achieved using only HDP.

 TRUE
 FALSE
6- True or False: Ambari is backed by RESTful APIs for developers to easily integrate with
their own applications.

 TRUE
 FALSE

7- Which Hadoop functionalities does Ambari provide?

 None of the above

 All of the above
 Monitor
 Manage
 Provision
 Integrate

8- Which page from the Ambari UI allows you to check the versions of the software installed
on your cluster?

 Monitor page
 Integrate page
 The Admin > Manage Ambari page
 The Admin > Provision page

9- True or False?Creating users through the Ambari UI will also create the user on the HDFS.

 TRUE
 FALSE

10- True or False? You can use the CURL commands to issue commands to Ambari.

 TRUE
 FALSE

11- True or False: Hadoop systems are designed for transaction processing.

 TRUE
 FALSE
12- What is the default number of replicas in a Hadoop system?

 5
 4
 3
 2

13- True or False: One of the driving principal of Hadoop is that the data is brought to the
program.

 TRUE
 FALSE

14- True or False: Atleast 2 Name Nodes are required for a standalone Hadoop cluster.

 TRUE
 FALSE

15- True or False: The phases in a MR job are Map, Shuffle, Reduce and Combiner

 TRUE
 FALSE
16- Centralized handling of job control flow is one of the the limitations of MR v1.

 TRUE
 FALSE

16- The Job Tracker in MR1 is replaced by which component(s) in YARN?

 ResourceMaster
 ApplicationMaster
 ApplicationManager
 ResourceManager

17- What are the benefits of using Spark?

(Please select the THREE that apply)

 Generality
 Versality
 Speed
 Ease of use
18- What are the languages supported by Spark?
(Please select the THREE that apply)

 Javascript
 HTML
 Python
 Java
 Scala

19- Resilient Distributed Dataset (RDD) is the primary abstraction of Spark.

 TRUE
 FALSE

20- What would you need to do in a Spark application that you would not need to do in a
Spark shell to start using Spark?

 Extract the necessary libraries to load the SparkContext

 Export the necessary libraries to load the SparkContext
 Delete the necessary libraries to load the SparkContext
 Import the necessary libraries to load the SparkContext

21- True or False: NoSQL database is designed for those that do not want to use SQL.

 TRUE
 FALSE

22- Which database is a columnar storage database?

 SQL
 Hive
 HBase

23- Which database provides a SQL for Hadoop interface?

 Hive
 Hadoop
 HBase
24- Which Apache project provides coordination of resources?

 Streams
 Spark
 Zeppelin
 ZooKeeper

25- What is ZooKeeper's role in the Hadoop infrastructure?

 Manage the coordination between HBase servers

 None of the above
 Hadoop and MapReduce uses ZooKeeper to aid in high availability of Resource
Manager
 All of the above
 Flume uses ZooKeeper for configuration purposes in recent releases

26- True or False: Slider provides an intuitive UI which allows you to dynamically allocate
YARN resources.

 TRUE
 FALSE

27- True or False: Knox can provide all the security you need within your Hadoop
infrastructure.

 TRUE
 FALSE

28- True or False: Sqoop is used to transfer data between Hadoop and relational databases.

 TRUE
 FALSE

29- True or False: For Sqoop to connect to a relational database, the JDBC JAR files for that
database must be located in $SQOOP_HOME/bin.

 TRUE
 FALSE
30- True or False: Each Flume node receives data as "source", stores it in a "channel", and
sends it via a "sink".

 TRUE
 FALSE

31- Through what HDP component are Kerberos, Knox, and Ranger managed?

 Zookeeper
 Ambari
 Apache Knox

32- Which security component is used to provide peripheral security?

 Apache Ranger
 Apache Camel
 Apache Knox

33- One of the governance issue that Hortonworks DataPlane Service (DPS) address is
visibility over all of an organization's data across all of their environments — on-prem,
cloud, hybrid — while making it easy to maintain consistent security and governance

 TRUE
 FALSE

34- True or false: The typical sources of streaming data are Sensors, "Data exhaust" and
high-rate transaction data.

 TRUE
 FALSE
35- What are the components of Hortonworks Data Flow(HDF)?

 Flow management
 Stream processing
 All of the above
 None of the above
 Enterprise services
36- True or False: NiFi is a disk-based, microbatch ETL tool that provides flow management

 TRUE
 FALSE

37- True or False: MiNiFi is a complementary data collection tool that feeds collected data to
NiFi

 TRUE
 FALSE

38- What main features does IBM Streams provide as a Streaming Data Platform?
(Please select the THREE that apply)

 Flow management
 Analysis and visualization
 Sensors
 Rich data connections
 Development support

39- What are the three types of Big Data?

(Please select the THREE that apply)

 Natural Language
 Semi-structured
 Graph-based
 Structured
 Machine-Generated
 Unstructured
40- What are the 4Vs of Big Data?
(Please select the FOUR that apply)

 Veracity
 Velocity
 Variety
 Value
 Volume
 Visualization

41- What are the most important computer languages for Data Analytics?
(Please select the THREE that apply)

 Scala
 HTML
 R
 SQL
 Python

42- True or False: GPUs are special-purpose processors that traditionally can be used to
power graphical displays, but for Data Analytics lend themselves to faster algorithm
execution because of the large number of independent processing cores.

 TRUE
 FALSE

43- True or False: Jupyter stores its workbooks in files with the .ipynb suffix. These files can
not be stored locally or on a hub server.

 TRUE
 FALSE

44- $BIGSQL_HOME/bin/bigsql startcommand is used to start Big SQL from the command line?

 TRUE
 FALSE
45- What are the two ways you can work with Big SQL.
(Please select the TWO that apply)

 JQuery
 R
 JSqsh
 Web tooling from DSM

46- What is one of the reasons to use Big SQL?

 Want to access your Hadoop data without using MapReduce

 You want to learn new languages like MapReduce
 Has deep learning curve because Big SQL uses standard 2011 query structure

47- Should you use the default STRING data type?

 Yes
 No

48- The BOOLEAN type is defined as SMALLINT SQL type in Big SQL.

 TRUE
 FALSE

49- Using the LOAD operation is the recommended method for getting data into your Big
SQL table for best performance.

 TRUE
 FALSE
50- Which file storage format has the highest performance?

 Delimited
 Sequence
 RC
 Parquet
 Avro
51- What are the two ways to classify functions?

 Built-in functions
 Scalar functions
 User-defined functions
 None of the above

52- True or False: UMASK is used to determine permissions on directories and files.

 TRUE
 FALSE
53- True or False: You can only Kerberize a Big SQL server before it is installed.

 TRUE
 FALSE

54- True or False: Authentication with Big SQL only occurs at the Big SQL layer or the client's
application layer.

 TRUE
 FALSE

55- True or False: Ranger and impersonation works well together.

 TRUE
 FALSE

56- True or False: RCAC can hide rows and columns.

 TRUE
 FALSE

57- True or False: Nicknames can be used for wrappers and servers.

 TRUE
 FALSE
58- True or False: Server objects defines the property and values of the connection.

 TRUE
 FALSE

59- True or False: The purpose of a wrapper provide a library of routines that doesn't
communicates with the data source.

 TRUE
 FALSE

60- True or False: User mappings are used to authenticate to the remote data source.

 TRUE
 FALSE
61- True or False: Collaboration with Watson Studio is an optional add-on component that
must be purchased.

 TRUE
 FALSE

62- True or False: Watson Studio is designed only for Data Scientists, other personas would
not know how to use it.

 TRUE
 FALSE

63- True or False: Community provides access to articles, tutorials, and even data sets that
you can use.

 TRUE
 FALSE

64- True or False: You can import visualization libraries into Watson Studio.

 TRUE
 FALSE
65- True or False: Collaborators can be given certain access levels.

 TRUE
 FALSE
66- True or False: Watson Studio contains Zeppelin as a notebook interface.

 TRUE
 FALSE
67- Spark is developed in which language

 Java
 Scala
 Python
 R

68- In Spark Streaming the data can be from what all sources?
 Kafka
 Flume
 Kinesis
 All of the above

69- Apache Spark has API's in

 Java
 Scala
 Python
 All of the above
70- Which of the following is not a component of Spark Ecosystem?
 Sqoop
 GraphX
 MLlib
 BlinkDB

S
71- Which is an advantage that Zeppelin holds over Jupyter?
Your answer

A. Notebooks can be used by multiple people at the same time.

B. Notebooks can be connected to big data engines such as Spark.

C. Users must authenticate before using a notebook.

D. Zeppelin is able to use the R language.

72- Why might a data scientist need a particular kind of GPU (graphics
processing unit)?
Your answer

A. To collect video for use in streaming data applications.

B. To display a simple bar chart of data on the screen.

C. To perform certain data transformation quickly.

D. To input commands to a data science notebook.

73- What command is used to list the "magic" commands in Jupyter?

Your answer

A. %list-all-magic

B. %dirmagic

C. %list-magic

D. %lsmagic
74- What is the first step in a data science pipeline?
Your answer

A. Exploration

B. Acquisition

C. Manipulation

D. Analytics

75- What is a markdown cell used for in a data science notebook?

Your answer

A. Documenting the computational process.

B. Configuring data connections.

C. Holding the output of a computation.

D. Writing code to transform data.

76- You have a distributed file system (DFS) and need to set permissions on
the the /hive/warehouse directory to allow access to ONLY the bigsql user.
Which command would you run?
Your answer

A. hdfs dfs -chmod 700 /hive/warehouse

B. hdfs dfs -chmod 755 /hive/warehouse

C. hdfs dfs -chmod 770 /hive/warehouse

D. hdfs dfs -chmod 666 /hive/warehouse

77- You need to determine the permission setting for a new schema directory.
Which tool would you use?
Your answer

A. umask

B. HDFS

C. Kerberos

D. GRANT

78- Which definition best describes RCAC?

Your answer

A. It grants or revokes certain user privileges.

B. It limits access by using views and stored procedures.

C. It grants or revokes certain directory privileges.

D. It limits the rows or columns returned based on certain criteria.

79- How many Big SQL management node do you need at minimum?
Your answer

A. 4

B. 1

C. 3

D. 2
80- Which directory permissions need to be set to allow all users to create
their own schema?
Your answer

A. 755

B. 666

C. 777

D. 700

81- Which command creates a user-defined schema function?

Your answer

A. CREATE FUNCTION

B. ALTER MODULE ADD FUNCTION

C. TRANSLATE FUNCTION

D. ALTER MODULE PUBLISH FUNCTION

82- What are Big SQL database tables organized into?

Your answer

A. Directories

B. Schemas

C. Hives

D. Files
83- Which Big SQL feature allows users to join a Hadoop data set to data in
external databases?
Your answer

A. Fluid query

B. Impersonation

C. Integration

D. Grant/Revoke privileges

84- Which two commands would you use to give or remove certain privileges
to/from a user?
Your answer

A. INSERT

B. GRANT

C. SELECT

D. REVOKE

E. LOAD
85- What is an advantage of the ORC file format?
Your answer

A. Efficient compression

B. Supported by multiple I/O engines

C. Data interchange outside Hadoop

D. Big SQL can exploit advanced features

86- You are creating a new table and need to format it with parquet. Which
partial SQL statement would create the table in parquet format?
Your answer

A. CREATE AS parquetfile

B. CREATE AS parquet

C. STORED AS parquetfile

D. STORED AS parquet

87- Which statement best describes a Big SQL database table?

Your answer

A. The defined format and rules around a delimited file.

B. A directory with zero or more data files.

C. A container for any record format.

D. A data type of a column describing its value.

88- You need to enable impersonation. Which two properties in the bigsql-
conf.xml file need to be marked true?
Your answer

A. bigsql.alltables.io.doAs

B. bigsql.impersonation.create.table.grant.public

C. DB2_ATS_ENABLE

D. DB2COMPOPT

E. $BIGSQL_HOME/conf

89- Using the Java SQL Shell, which command will connect to a database
called mybigdata?
Your answer

A. ./java tables

B. ./jsqsh mybigdata

C. ./java mybigdata

D. ./jsqsh go mybigdata

90- When connecting to an external database in a federation, you need to use

the correct database driver and protocol. What is this federation component
called in Big SQL?
Your answer
A. Wrapper

B. Data source

C. Nickname

D. User mapping

91- What are two primary limitations of MapReduce v1?

Your answer

A. Scalability

B. Workloads limited to MapReduce

C. Resource utilization

D. TaskTrackers can be a bottleneck to MapReduce jobs

E. Number of TaskTrackers limited to 1,000

92- Which feature makes Apache Spark much easier to use than MapReduce?
Your answer

A. Suitable for transaction processing.

B. Libraries that support SQL queries.

C. Applications run in-memory.

D. APIs for Scala, Python, C++, and .NET.

93- What is an example of a NoSQL datastore of the "Document Store" type?

Your answer

A. Cassandra

B. REDIS

C. HBase
D. MongoDB

94- Which Apache Hadoop application provides a high-level programming

language for data transformation on unstructured data?
Your answer

A. Zookeeper

B. Pig

C. Hive

D. Sqoop

95- Under the MapReduce v1 programming model, which shows the proper
order of the full set of MapReduce phases?
Your answer

A. Map -> Split -> Reduce -> Combine

B. Map -> Combine -> Reduce -> Shuffle

C. Map -> Combine -> Shuffle -> Reduce

D. Split -> Map -> Combine -> Reduce

96- Which three programming languages are directly supported by Apache

Spark?
Your answer

A. C#

B. C++

C. Java

D. .NET
E. Python

F. Scala

97- Which statement accurately describes how ZooKeeper works?

Your answer

A. All servers keep a copy of the shared data in memory.

B. Clients connect to multiple servers at the same time.

C. There can be more than one leader server at a time.

D. Writes to a leader server will always succeed.

98- Which computing technology provides Hadoop's high performance?

Your answer

A. Online Transactional Processing

B. Parallel Processing

C. Online Analytical Processing

D. RAID-0

99- Which two factors in a Hadoop cluster increase performance most

significantly?
Your answer

A. immediate failover of failed disks

B. parallel reading of large data files

C. data redundancy on management nodes

D. solid state disks

E. high-speed networking between nodes

F. large number of small data files

100- Which component of the Apache Ambari architecture provides statistical

data to the dashboard about the performance of a Hadoop cluster?
Your answer

A. Ambari Alert Framework

B. Ambari Wizard

C. Ambari Metrics System

D. Ambari Server

101- Apache Spark can run on which two of the following cluster managers?
Your answer

A. oneSIS

B. Linux Cluster Manager

C. Nomad

D. Apache Mesos

E. Hadoop YARN

102- Under the MapReduce v1 programming model, which optional phase is

executed simultaneously with the Shuffle phase?
Your answer

A. Map

B. Combiner

C. Reduce
D. Split

103- Which hardware feature on an Hadoop datanode is recommended for cost

efficient performance?
Your answer

A. JBOD

B. RAID

C. LVM

D. SSD

104- What is the name of the Hadoop-related Apache project that utilizes an in-
memory architecture to run applications faster than MapReduce?
Your answer

A. Spark

B. Python

C. Pig

D. Hive

105- Which statement is true about MapReduce v1 APIs?

Your answer

A. MapReduce v1 APIs cannot be used with YARN.

B. MapReduce v1 APIs are implemented by applications which are largely

independent of the execution environment.

C. MapReduce v1 APIs define how MapReduce jobs are executed.

D. MapReduce v1 APIs provide a flexible execution environment to run
MapReduce.

106- Hadoop uses which two Google technologies as its foundation?

Your answer

A. YARN

B. Google File System

C. Ambari

D. HBase

E. MapReduce

107- What are two common issues in distributed systems?

Your answer

A. Reduced performance when compared to a single server.

B. Partial failure of the nodes during execution.

C. Finding a particular node within the cluster.

D. Distributed systems are harder to scale up.

108- Which statement about Apache Spark is true?

Your answer

A. It runs on Hadoop clusters with RAM drives configured on each

DataNode.

B. It is much faster than MapReduce for complex applications on disk.

C. It supports HDFS, MS-SQL, and Oracle.

D. It features APIs for C++ and .NET.

109- Which two are valid watches for ZNodes in ZooKeeper?
Your answer

A. NodeRefreshed

B. NodeExpired

C. NodeChildrenChanged

D. NodeDeleted

110- Which Apache Hadoop component can potentially replace an RDBMS as a

large Hadoop datastore and is particularly good for "sparse data"?
Your answer

A. Spark

B. Ambari

C. HBase

D. MapReduce

111- Which statement describes an example of an application using streaming

data?
Your answer

A. An application evaluating sensor data in real-time.

B. One time export and import of a database.

C. A web application that supports 10,000 users.

D. A system that stores many records in a database.

112- Which component of the Hortonworks Data Platform (HDP) is the
architectural center of Hadoop and provides resource management and a
central platform for Hadoop applications?
Your answer

A. HBase

B. HDFS

C. YARN

D. MapReduce

113- Which three are a part of the Five Pillars of Security?

Your answer

A. Administration

B. Audit

C. Resiliency

D. Speed

E. Data Protection

114- How can a Sqoop invocation be constrained to only run one mapper?
Your answer

A. Use the --limit mapper=1 parameter.

B. Use the -mapper 1 parameter.

C. Use the --single parameter.

D. Use the -m 1 parameter.

115- Under the YARN/MRv2 framework, which daemon is tasked with
negotiating with the NodeManager(s) to execute and monitor tasks?
Your answer

A. TaskManager

B. JobMaster

C. ResourceManager

D. ApplicationMaster

116- Apache Spark provides a single, unifying platform for which three of the
following types of operations?
Your answer

A. ACID transactions

B. graph operations

C. record locking

D. batch processing

E. machine learning

F. transaction processing

117- Which Apache Hadoop application provides an SQL-like interface to allow

abstraction of data on semi-structured data in a Hadoop datastore?
Your answer

A. Spark

B. Pig

C. YARN

D. Hive
118- Under the MapReduce v1 programming model, what happens in a
"Reduce" step?
Your answer

A. Data is aggregated by worker nodes.

B. Worker nodes process pieces in parallel.

C. Worker nodes store results on their own local file systems.

D. Input is split into pieces.

119- What are two security features Apache Ranger provides?

Your answer

A. Auditing

B. Authentication

C. Authorization

D. Availability

120- Under the YARN/MRv2 framework, the JobTracker functions are split into
which two daemons?
Your answer

A. JobMaster

B. TaskManager

C. ApplicationMaster

D. ScheduleManager

E. ResourceManager
121- Under the YARN/MRv2 framework, which daemon arbitrates the execution
of tasks among all the applications in the system?
Your answer

A. ApplicationMaster

B. JobMaster

C. ScheduleManager

D. ResourceManager

122- Which description characterizes a function provided by Apache Ambari?

Your answer

A. A wizard for installing Hadoop services on host servers.

B. Moves large amounts of streaming event data.

C. Moves information to/from structured databases.

D. A messaging system for real-time data pipelines.

123- What is the preferred replacement for Flume?

Your answer

A. Druid

B. Hortonworks Data Flow

C. NiFi

D. Storm
124- If a Hadoop node goes down, which Ambari component will notify the
Administrator?
Your answer

A. Ambari Metrics System

B. Ambari Alert Framework

C. Ambari Wizard

D. REST API

125- Hadoop 2 consists of which three open-source sub-projects maintained

by the Apache Software Foundation?
Your answer

A. HDFS

B. Hive

C. YARN

D. MapReduce

E. Big SQL

F. Cloudbreak

126- What is the architecture of Watson Studio centered on?

Your answer

A. Projects

B. Data Assets

C. Analytic Assets

D. Collaborators
127- Which type of cell can be used to document and comment on a process in
a Jupyter notebook?
Your answer

A. Markdown

B. Code

C. Kernel

D. Output

128- Which Watson Studio offering used to be available through something

known as IBM Bluemix?
Your answer

A. Watson Studio Business

B. Watson Studio Local

C. Watson Studio Cloud

D. Watson Studio Desktop

129- Where does the unstructured data of a project reside in Watson Studio?
Your answer

A. Wrapper

B. Database

C. Object Storage

D. Tables
130- Before you create a Jupyter notebook in Watson Studio, which two items
are necessary?
Your answer

A. Spark Instance

B. Scala

C. Project

D. File

E. URL

130- The basic abstraction of Spark Streaming is

 Dstream
 RDD
 Shared Variable
 None of the above

131- Which of the following algorithm is not present in MLlib?

 Streaming Linear Regression

 Streaming KMeans
 Tanimoto distance
 None of the above

132- Dstream internally is

 Continuous Stream of RDD

 Continuous Stream of DataFrame
 Continuous Stream of DataSet
 None of the above

133- Can we add or setup new string computation after SparkContext starts

 Yes
 No
134- Which of the following is not the feature of Spark?

 Supports in-memory computation

 Fault-tolerance
 It is cost efficient
 Compatible with other file storage system

135- Which is the abstraction of Apache Spark?

 Shared Variable
 RDD
 Both the above

136- What are the parameters defined to specify window operation

 Window length, sliding interval
 State size, window length
 State size, sliding interval
 None of the above

137- Which of the following is not output operation on DStream

 SaveAsTextFiles
 ForeachRDD
 SaveAsHadoopFiles
 ReduceByKeyAndWindow

138- Dataset was introduced in which Spark release?

 Spark 1.6
 Spark 1.4.0
 Spark 2.1.0
 Spark 1.1

139- Which Cluster Manager do Spark Support?

 Standalone Cluster Manager
 MESOS
 YARN
 All of the above

140- The default storage level of cache() is?

 MEMORY_ONLY
 MEMORY_AND_DISK
 DISK_ONLY
  MEMORY_ONLY_SER

141- Which is not a component on the top of Spark Core?

 Spark RDD
 Spark Streaming
 MLlib
 None of the above

142- Apache Spark was made open-source in which year?

 2010
 2011
 2008
 2009

143- In addition to stream processing jobs, what all functionality do Spark provides?
 Machine learning
 Graph processing
 Batch processing
 All of the above

144- Is Spark included in every major distribution of Hadoop?

 Yes
 No

145- Which of the following is not true for Hadoop and Spark?
 Both are data processing platforms
 Both are cluster computing environments
 Both have their own file system
 Both use open source APIs to link between different tools

146- How much faster can Apache Spark potentially run batch-processing programs when
processed in memory than MapReduce can?
 10 times faster
 20 times faster
 100 times faster
 200 times faster
147- Which of the following provide the Spark Core’s fast scheduling capability to perform
streaming analytics.
 RDD
 GraphX
 Spark Streaming
 Spark R

148- Which of the following is the reason for Spark being Speedy than MapReduce?
 DAG execution engine and in-memory computation
 Support for different language APIs like Scala, Java, Python and R
 RDDs are immutable and fault-tolerant
 None of the above

149- Can you combine the libraries of Apache Spark into the same Application, for example,
MLlib, GraphX, SQL and DataFrames etc.
 Yes
 No

150- Which of the following is true for RDD?

 RDD is programming paradigm
 RDD in Apache Spark is an immutable collection of objects
 It is database
 None of the above

151- Which of the following is not a function of Spark Context in Apache Spark?
 Entry point to Spark SQL
 To Access various services
 To set the configuration
 To get the current status of Spark Application

152- What are the features of Spark RDD?

 In-memory computation
 Lazy evaluations
 Fault Tolerance
 All of the above

153- How many Spark Context can be active per JVM?

 More than one
 Only one
 Not specific
 None of the above
154- In how many ways RDD can be created?
 4
 3
 2
 1

155- How many tasks does Spark run on each partition?

 Any number of task
 One
 More than one less than five

156- Can we edit the data of RDD, for example, the case conversion?
 Yes
 No

157- Which of the following is not a transformation?

 Flatmap
 Map
 Reduce
 Filter

158- Which of the following is not an action?

 collect()
 take(n)
 top()
 map

159- Does Spark R make use of MLlib in any aspect?

 Yes
 No

160- You can connect R program to a Spark cluster from -

 RStudio
 R Shell
 Rscript
 All of the above

161- For Multiclass classification problem which algorithm is not the solution?
 Naive Bayes
 Random Forests
  Logistic Regression
 Decision Trees

162- For Regression problem which algorithm is not the solution?

 Logistic Regression
 Ridge Regression
 Decision Trees
 Gradient-Boosted Trees

163- Which of the following is true about DataFrame?

 DataFrames provide a more user-friendly API than RDDs.
 DataFrame API have provision for compile-time type safety
 Both the above

164- Which of the following is a tool of Machine Learning Library?

 Persistence
 Utilities like linear algebra, statistics
 Pipelines
 All of the above

165- Is MLlib deprecated?

 Yes
 No

166- Which of the following is false for Apache Spark?

 It provides high-level API in Java, Python, R, Scala
 It can be integrated with Hadoop and can process existing Hadoop HDFS data
 Spark is an open source framework which is written in Java
 Spark is 100 times faster than Bigdata Hadoop

167- Which of the following is true for Spark SQL?

 It is the kernel of Spark
 Provides an execution platform for all the Spark applications
 It enables users to run SQL / HQL queries on the top of Spark.
 Enables powerful interactive and data analytics application across live streaming
data

168- Which of the following is true for Spark core?

 It is the kernel of Spark
 It enables users to run SQL / HQL queries on the top of Spark.
 It is the scalable machine learning library which delivers efficiencies
 Improves the performance of iterative algorithm drastically.
169- Which of the following is true for Spark R?
 It allows data scientists to analyze large datasets and interactively run jobs
 It is the kernel of Spark
 It is the scalable machine learning library which delivers efficiencies
 It enables users to run SQL / HQL queries on the top of Spark.

170- Which of the following is true for Spark MLlib?

 Provides an execution platform for all the Spark applications
 It is the scalable machine learning library which delivers efficiencies
 enables powerful interactive and data analytics application across live streaming
data
 All of the above

171- Which of the following is true for Spark Shell?

 It helps Spark applications to easily run on the command line of the system
 It runs/tests application code interactively
 It allows reading from many types of data sources
 All of the above

172- Which of the following is true for RDD?

 We can operate Spark RDDs in parallel with a low-level API
 RDDs are similar to the table in a relational database
 It allows processing of a large amount of structured data
 It has built-in optimization engine

173- RDD are fault-tolerant and immutable

 True
 False

174- In which of the following cases do we keep the data in-memory?

 Iterative algorithms
 Interactive data mining tools
 Both the above

175- When does Apache Spark evaluate RDD?

 Upon action
 Upon transformation
 On both transformation and action

176- The read operation on RDD is

 Fine-grained
  Coarse-grained
 Either fine-grained or coarse-grained
 Neither fine-grained nor coarse-grained

177- The write operation on RDD is

 Fine-grained
 Coarse-grained
 Either fine-grained or coarse-grained
 Neither fine-grained nor coarse-grained

178- Is it possible to mitigate stragglers in RDD?

 Yes
 No

179- Fault Tolerance in RDD is achieved using

 Immutable nature of RDD
 DAG (Directed Acyclic Graph)
 Lazy-evaluation
 None of the above

180- What is a transformation in Spark RDD?

 Takes RDD as input and produces one or more RDD as output.
 Returns final result of RDD computations.
 The ways to send result from executors to the driver
 None of the above

181- What is action in Spark RDD?

 The ways to send result from executors to the driver
 Takes RDD as input and produces one or more RDD as output.
 Creates one or many new RDDs
 All of the above

182- Which of the following is true about narrow transformation -

 The data required to compute resides on multiple partitions.
 The data required to compute resides on the single partition.
 Both the above

183- Which of the following is true about wide transformation -

 The data required to compute resides on multiple partitions.
 The data required to compute resides on the single partition.
 None of the both
184-Whenwe want to work with the actual dataset, at that point we use Transformation?
 True
 False

185- The shortcomings of Hadoop MapReduce was overcome by Spark RDD by

 Lazy-evaluation
 DAG
 In-memory processing
 All of the above

186- What does Spark Engine do?

 Scheduling
 Distributing data across a cluster
 Monitoring data across a cluster
 All of the above

187- Caching is optimizing the technique

 True
 False

188- Which of the following is the entry point of Spark Application -

 SparkSession
 SparkContext
 None of the both

189- SparkContext guides how to access the Spark cluster.

 True
 False

190- Which of the following is the entry point of Spark SQL?

 SparkSession
 SparkContext

191- Which of the following is open-source?

 Apache Spark
 Apache Hadoop
 Apache Flink
 All of the above
192- Apache Spark supports -
 Batch processing
 Stream processing
 Graph processing
 All of the above

193- Which of the following is not true for map() Operation?

 Map transforms an RDD of length N into another RDD of length N.
 In the Map operation developer can define his own custom business logic.
 It applies to each element of RDD and it returns the result as new RDD
 Map allows returning 0, 1 or more elements from map function.

194- FlatMap transforms an RDD of length N into another RDD of length M. which of the
following is true for N and M.

a. N>M

b. N<M

c. N<=M

 Either a or b
 Either b or c
 Either a or c

195- Which of the following is a transformation?

 take(n)
 top()
 countByValue()
 mapPartitionWithIndex()

196- Which of the following is action?

 Union(dataset)
 Intersection(other-dataset)
 Distinct()
 CountByValue()

197- In aggregate function can we get the data type different from as that input data type?
 Yes
 No
198- In which of the following Action the result is not returned to the driver.
 collect()
 top()
 countByValue()
 foreach()

199- Which of the following is true for stateless transformation?

 Uses data or intermediate results from previous batches and computes the result of
the current batch.
 Windowed operations and updateStateByKey() are two type of Stateless
transformation.
 The processing of each batch has no dependency on the data of previous batches.
 None of the above

120- Which of the following is true for stateful transformation?

 The processing of each batch has no dependency on the data of previous batches.
 Uses data or intermediate results from previous batches and computes the result of
the current batch.
 Stateful transformations are simple RDD transformations.
 None of the above

121- The primary Machine Learning API for Spark is now the _____ based API
 DataFrame
 Dataset
 RDD
 All of the above

122- Which of the following is a module for Structured data processing?

 GraphX
 MLlib
 Spark SQL
 Spark R

123- SparkSQL translates commands into codes. These codes are processed by
 Driver nodes
 Executor Nodes
 Cluster manager
 None of the above

124- Spark SQL plays the main role in the optimization of queries.
 True
  False

125- This optimizer is based on functional programming construct in

 Java
 Scala
 Python
 R

126- Catalyst Optimizer supports either rule-based or cost-based optimization.

 True
 False

127- Which of the following is not true for Catalyst Optimizer?

 Catalyst optimizer makes use of pattern matching feature.
 Catalyst contains the tree and the set of rulesto manipulate the tree.
 There are no specific libraries to process relational queries.
 There are different rule sets which handle different phases of query.

128- Which of the following is true for the tree in Catalyst optimizer?
 A tree is the main data type in the catalyst.
 New nodes are defined as subclasses of TreeNode class.
 A tree contains a node object.
 All of the above

129- Which of the following is true for the rule in Catalyst optimizer?
 We can manipulate tree using rules.
 We can define rules as a function from one tree to another tree.
 Using rule we get the pattern that matches each pattern to a result.
 All of the above

130- Which of the following is not a Spark SQL query execution phases?
 Analysis
 Logical Optimization
 Execution
 Physical planning

131- In Spark SQL optimization which of the following is not present in the logical plan -
 Constant folding
 Abstract syntax tree
 Projection pruning
 Predicate pushdown
132- In the analysis phase which is the correct order of execution after forming unresolved
logical plan

a. Search relation BY NAME FROM CATALOG.

b. Determine which attributes match to the same value to give them unique ID.
c. Map the name attribute
d. Propagate and push type through expressions

 abcd
 acbd
 adbc
 dcab

133-In the Physical planning phase of Query optimization we can use both Coast-based and
Rule-based optimization.
 True
 False

134-DataFramein Apache Spark prevails over RDD and does not contain any feature of RDD.
 True
 False

135- Which of the following are the common feature of RDD and DataFrame?
 Immutability
 In-memory
 Resilient
 All of the above

136- Which of the following is not true for DataFrame?

 DataFrame in Apache Spark is behind RDD
 We can build DataFrame from different data sources. structured data file, tables in
Hive
 The Application Programming Interface (APIs) of DataFrame is available in various
languages
 Both in Scala and Java, we represent DataFrame as Dataset of rows.

137- In Dataframe in Spark Once the domain object is converted into a data frame, the
regeneration of domain object is not possible.
 True
 False
138- DataFrameAPI has provision for compile-time type safety.
 True
 False

139- We can create DataFrame using:

 Tables in Hive
 Structured data files
 External databases
 All of the above

140- Which of the following is the fundamental data structure of Spark

 RDD
 DataFrame
 Dataset
 None of the above

141- Which of the following organized a data into a named column?

a. RDD
b. DataFrame
c. Dataset

 Both a and b
 Both b and c
 Both a and c

142- Which of the following provide the object-oriented programming interface

 RDD
 DataFrame
 Dataset
 None of the above

143- After transforming into DataFrame one cannot regenerate a domain object
 True
 False

144- RDD allows Java serialization

 True
 False
145- Which of the following make use of an encoder for serialization.
 RDD
 DataFrame
 Dataset
 None of the above

146- Apache Spark is presently added in all major distribution of Hadoop

 True
 False

147-DoesDataset API support Python and R.

 Yes
 No

148- Which of the following is slow to perform simple grouping and aggregation operations.
 RDD
 DataFrame
 Dataset
 All of the above

149- Which of the following is good for low-level transformation and actions.
 RDD
 DataFrame
 Dataset
 All of the above

150- Which of the following technology is good for Stream technology?

 Apache Spark
 Apache Hadoop
 Apache Flink
 None of the above

151- Which of the following is not true for Apache Spark Execution?
 To simplify working with structured data it provides DataFrame abstraction in
Python, Java, and Scala.
 The data can be read and written in a variety of structured formats. For example,
JSON, Hive Tables, and Parquet.
 Using SQL we can query data,only from inside a Spark program and not from
external tools.
 The best way to use Spark SQL is inside a Spark application. This empowers us to
load data and query it with SQL.
152- When SQL run from the other programming language the result will be
 DataFrame
 DataSet
 Either DataFrame or Dataset
 Neither DataFrame nor Dataset

153- The Dataset API is accessible in

 Java and Scala
 Java, Scala and python
 Scala and Python
 Scala and R

154- Dataset API is not supported by Python. But because of the dynamic nature of Python,
many benefits of Dataset API are available.
 True
 False
155- Which of the following is true for Catalyst optimizer?
 The optimizer helps us to run queries much faster than their counter RDD part.
 The optimizer helps us to run queries little faster than their counter RDD part.
 The optimizer helps us to run queries in the same speed as their counter RDD part.

156- Which of the following are uses of Apache Spark SQL?

 It executes SQL queries.
 We can read data from existing Hive installation using SparkSQL.
 When we run SQL within another programming language we will get the result as
Dataset/DataFrame.
 All of the above

157- With the help of Spark SQL, we can query structured data as a distributed dataset
(RDD).
 True
 False

158- Spark SQL can connect through JDBC or ODBC.

 True
 False
159- Using Spark SQL, we can create or read a table containing union fields.
 True
 False
160- Which of the following is true for Spark SQL?
 Hive transactions are not supported by Spark SQL.
 No support for time-stamp in Avro table.
 Even if the inserted value exceeds the size limit, no error will occur.
 All of the above

161- Which of the following is the daemon of Hadoop?

 NameNode
 Node manager
 DataNode
 All of the above

162-Which one of the following is false about Hadoop?

 It is a distributed framework
 The main algorithm used in it is Map Reduce
 c. It runs with commodity hard ware
 All are true

163-Hadoop Framework is written in

 Python
 Java
 C++
 Scala

164- Whichof the following is component of Hadoop?

 YARN
 HDFS
 MapReduce
 All of the above

165- The archive file created in Hadoop has the extension of

 .hrh
 .har
 .hrc
 .hrar

166- Whichcommand is used to check the status of all daemons running in the HDFS.
 jps
 fsck
 distcp
 None of the above
167- What license is Apache Hadoop distributed under?
 Apache License 2.0
 Shareware
 Mozilla Public License
 Commercial

168- Which of the following platforms does Apache Hadoop run on ?

 Bare metal
 Unix-like
 Cross-platform
 Debian

169- Apache Hadoop achieves reliability by replicating the data across multiple hosts, and
hence does not require ________ storage on hosts.
 Standard RAID levels
 RAID
 ZFS
 Operating system

170- Which of the following is the correct statement

 Data locality means moving computation to data instead of data to computation
 Data locality means moving data to computation instead of computation to data
 Both the above
 None of the above

171- Hadoop works in

 master-worker fashion
 master – slave fashion
 worker/slave fashion
 All of the mentioned

172- Which type of data Hadoop can deal with is

 Structured
 Semi - structured
 Unstructured
 All of the above

173- Which statement is false about Hadoop

 It runs with commodity hard ware
 It is a part of the Apache project sponsored by the ASF
 It is best for live streaming of data
  None of the above

174- Whichof the following property gets configured on mapred-site.xml ?

 Replication factor
 Java Environment variables.
 Directory names to store hdfs files.
 Host and port where MapReduce job runs.

175- Which of the below apache system deals with ingesting streaming data to hadoop
 Flume
 Oozie
 Hive
 Kafka

176- As compared to RDBMS, Apache Hadoop

 Has higher data Integrity
 Does ACID transactions
 Is suitable for read and write many times
 Works better on unstructured and semi-structured data.

177- Whichcommand lists the blocks that make up each file in the filesystem
 hdfs fsck / -files -blocks
 hdfs fsck / -blocks -files
 hdfs fchk / -blocks -files
 hdfs fchk / -files -blocks

178- In which all languages you can code in Hadoop

 Java
 Python
 C++
 All of the above

179- Whchof the file contains the configuration setting for HDFS daemons
 yarn-site.xml
 hdfs-site.xml
 mapred-site.xml
 None of the above

180- All of the following accurately describe Hadoop, EXCEPT

 Open source
 Real-time
 Java-based
  Distributed computing approach

181- Whchof the file contains the configuration setting for NodeManager and
ResourceManager
 yarn-site.xml
 hdfs-site.xml
 mapred-site.xml
 None of the above

182- Hadoop can be used to create distributed clusters, based on commodity servers, that
provide low-cost processing and storage for unstructured data
 True
 False

183- Which of the following is a distributed multi-level database?

 HDFS
 HBase
 Both the above
 None of the above

184- Which of the following is used for machine learning on Hadoop?

 Hive
 Pig
 HBase
 Mahoot

185- Which of the following is used to ingest streaming data into Hadoop clusters
 Flume
 Sqoop
 Both the above
 None of the above

186- Hadoop distributed file system behaves similarly to which of the following:
 RAID-1 Filesystem
 RAID-0 Filesystem
 Both the above
 All of the above

187- Zookeeper ensures that

 All the namenodes are actively serving the client requests
 A failover is triggered when any of the datanode fails.
 Only one namenode is actively serving the client requests
  A failover can not be started by hadoop administrator.

188- Which of the following is used to ingest data into Hadoop clusters?
 Flume
 Sqoop
 Both the above
 Nonw of the above

189- Which of the following is a data processing engine for clustered computing?
 Drill
 Oozie
 Spark
 All of the above

190- Whichtool could be used to move data from RDBMS data to HDFS?
 Sqoop
 Flume
 Both the above
 None of the above

191- All the files in a directory in HDFS can be merged together using which of the
following?
 Put merge
 Get merge
 Remerge
 Merge all

192- Whichof these provides a Stream processing system used in Hadoop ecosystem?
 Hive
 Solr
 Tez
 Spark

193- The client reading the data from HDFS filesystem in Hadoop does which of the
following?
 Gets only the block locations form the namenode
 Gets the data from the namenode
 Gets both the data and block location from the namenode
 Gets the block location from the datanode

194- Whichof the following jobs are optimized for scalability but not latency
 Mapreduce
  Drill
 Oozie
 Hive

195- Can multiple clients write into an HDFS file concurrently?

 Yes
 No

196- Which of the following command is used to enter Safemode

 hadoop dfsadmin –safemode get
 bin dfsadmin –safemode get
 hadoop dfsadmin –safemode enter
 None of the above

197- Whichof the following command is used to come out of Safemode

 hadoop dfsadmin -safemode leave
 hadoop dfsadmin -safemode exit
 hadoop dfsadmin -safemode out
 None of the above

198- DuringSafemode Hadoop cluster is in

 Read-only
 Write-only
 Read-Write
 None of the above

199- HDFS allows a client to read a file which is already opened for writing?
 False
 True

200- What happens when a file is deleted from the command line?
 It is permanently deleted if trash is enabled.
 It is permanently deleted and the file attributes are recorded in a log file.
 It is placed into a trash directory common to all users for that cluster.
 None of the above

201- Whichone of the following statements is false about Distributed Cache?

 Hive
 Drill
 Oozie
 Mapreduce
202- Which among the following are the features of Hadoop
 Open source
 Fault-tolerant
 High Availability
 All of the above

203- Checkpoint node download the FsImage and EditLogs from the NameNode & then
merge them & store the modified FsImage
 Into persistent storage
 Back to the active NameNode

204- Whichstatement is true about Safemode

 It is a maintenance state of NameNode
 In Safemode, HDFS cluster is in read only
 In Safemode, NameNode doesn’t allow any modifications to the file system
 All of the above

205- Which command is used to know the current status of the safe mode
 hadoop dfsadmin –safemode get
 hadoop dfsadmin –safemode getStatus
 hadoop dfsadmin –safemode status
 None of the above

206- Which of the following deal with small files issue

 Hadoop archives
 Sequence files
 HBase
 All of the above

207- Whichof the following feature overcomes this single point of failure
 None of the above
 HDFS federation
 High availability
 Erasure coding

208- Which statement is true about NameNode High Availability

 Solve Single point of failure
 For high scalability
 Reduce storage overhead to 50%
 None of the above
209- In NameNode HA, when active node fails, which node takes the responsibility of active
node
 Secondary NameNode
 Backup node
 Standby node
 Checkpoint node

210- Whatare the advantages of 3x replication schema in Hadoop

 Fault tolerance
 High availability
 Reliability
 All of the above

211- Whatare the advantages of HDFS federation in Hadoop?

 Isolation
 Namespace scalability
 Improves throughput
 All of the above

212- Rack Awareness improves

 Data high availability and reliability
 Performance of the cluster
 Network bandwidth
 All of the above

213- Which property is used to enable/disable speculative execution

 mapred.map.tasks.speculative.execution
 mapred.reduce.tasks.speculative.execution
 Both the above
 None of the above

214- In which process duplicate task is created to improve the overall execution time
 Erasure coding
 Speculative execution
 HDFS federation
 None of the above

215- In which mode each daemon runs on a single node but there is separate java process
for each daemon
 Local (Standalone) mode
 Fully distributed mode
  Pseudo-distributed mode
 None of the above

216- In which mode each daemon runs on a single node as a single java process
 Local (Standalone) mode
 Pseudo-distributed mode
 Fully distributed mode
 None of the above

217- In which mode all daemons execute in separate nodes

 Local (Standalone) mode
 Pseudo-distributed mode
 Fully distributed mode
 None of the above

218- Which configuration file is used to control the HDFS replication factor?
 mapred-site.xml
 hdfs-site.xml
 core-site.xml
 yarn-site.xml

219- How to adjust the size of a distributed cache

 local.cache.size.
 mapred.cache.size.
 hdfs.cache.size.
 distributedcache.size.

220- Whatis the default size of distributed cache

 8 GB
 10 GB
 12 GB
 16 GB

221- Distributed cache can cache files

 Jar Files
 Read-only text files
 Archives
 All of the above

222- The total number of partitioner is equal to

 The number of reducer
 The number of mapper
  The number of combiner
 None of the above

223- Which of the following Hadoop config files is used to define the heap size?
 hdfs-site.xml
 core-site.xml
 hadoop-env.sh
 mapred-site.xml

224- Which of the following feature you will use submit jars, static files for MapReduce job
during runtime
 Distributed cache
 Speculative execution
 Data locality
 Erasure coding

225- Which of the following method used to set the output directory
 FileOutputFormat.setOutputgetpath()
 OutputFormat.setOutputpath()
 FileOutputFormat.setOutputpath()
 OutputFormat.setOutputgetpath()

226- Which tool is used to distributes data evenly across datanode

 Balancer
 Disk balancer

227- Which tool is used to distributes data evenly on all disks of a datanode
 Balancer
 Disk Balancer

228- Which of the following must be set true enable diskbalnecr in hdfs-site.xml
 dfs.balancer.enabled
 dfs.disk.balancer.enabled
 dfs.diskbalancer.enabled
 dfs.disk.balancer.enabled

229- In disk balancer datanode uses which volume choosing the policy to choose the disk for
the block.
 Round-robin
 Available space
 All of the above
 None of the above
230- Which among the following is configuration files in Hadoop
 core-site.xml
 hdfs-site.xml
 yarn-site.xml
 All of the above

231- Which of the following is used for large inter/intra-cluster copying

 fsck
 distch
 DistCp
 dtutil

232- Hadoop uses hadoop

 Troubleshooting
 Performance reporting purpose
 Monitoring
 All of the above

233- Is it possible to provide multiple inputs to Hadoop?

 Yes
 No

234- Is it possible to have Hadoop job output in multiple directories?

 Yes
 No

235- Which of the following is used to provide multiple outputs to Hadoop?

 MultipleOutputFormat
 MultipleOutputs class
 FileOutputFormat
 DBInputFormat

236- Which of the following command is used to check for various inconsistencies
 zkfc
 fs
 fsck
 fetchdt

237- What does commodity Hardware in Hadoop world mean?

 Very cheap hardware
  Industry standard hardware
 Discarded hardware
 Low specifications Industry grade hardware

238- Distributed cache files can’t be accessed in Reducer.

 True
 False

239- Pig is a:
 Programming Language
 Data Flow Language
 Query Language
 Database

240- Distributed Cache can be used in

 Mapper phase only
 Reducer phase only
 In either phase, but not on both sides simultaneously
 In either phase

241- Which of the following is not a valid Hadoop config file?

 core-default.xml
 hdfs-default.xml
 hadoop-default.xml
 mapred-default.xml

242- It is necessary to default all the properties in Hadoop config files.

 True
 False

243- Which of the following is a column-oriented database that runs on top of HDFS
 Hive
 Sqoop
 HBase
 Flume

244- Which command is used to show all the Hadoop daemons that are running on the
machine
 distcp
 jps
 fsck
245- Hadoop is a framework that works with a variety of related tools. Common cohorts
include:
 MapReduce, Hive and HBase
 MapReduce, MySQL and Google Apps
 MapReduce, Hummer and Iguana
 MapReduce, Heron and Trumpet
Question: 1
Which capability does IBM BigInsights add to enrich
Hadoop?
Your answer

A Jaql

B Fault tolerance through HDFS replication

C Adaptive MapReduce

D Parallel computing on commodity servers

Question: 2
What is one of the four characteristics of Big Data?
Your answer

A value

B volume

C verifiability

D volatility

Question: 3
Which Hadoop-related project provides common
utilities and libraries that support other Hadoop sub
projects?
Your answer

A Hadoop Common
Your answer

B Hadoop HBase

C MapReduce

D BigTable

Question: 4
Which type of Big Data analysis involves the
processing of extremely large volumes of constantly
moving data that is impractical to store?
Your answer

A Federated Discovery and Navigation

B Text Analysis

C Stream Computing

D MapReduce

Question: 6
Which primary computing bottleneck of modern
computers is addressed by Hadoop?
Your answer

A 64-bit architecture

B disk latency

C MIPS
Your answer

D limited disk capacity

Question: 7
Which Big Data function improves the decision-
making capabilities of organizations by enabling the
organizations to interpret and evaluate structured
and unstructured data in search of valuable
business information?
Your answer

A stream computing

B data warehousing

C analytics

D distributed file system

Question: 8
What is one of the two technologies that Hadoop
uses as its foundation?
Your answer

A HBase

B Apache

C Jaql

D MapReduce
Question: 9
What key feature does HDFS 2.0 provide that HDFS
does not?
Your answer

A a high throughput, shared file system

B high availability of the NameNode

C data access performed by an RDBMS

D random access to data in the cluster

Question: 10
What are two of the core operators that can be used
in a Jaql query? (Select two.)
Your answer

A LOAD

B JOIN

C TOP

D SELECT

Question: 11
Which type of language is Pig?
Your answer

A SQL-like

B compiled language
Your answer

C object oriented

D data flow

Question: 12
If you need to change the replication factor or
increase the default storage block size, which file do
you need to modify?
Your answer

A hdfs.conf

B hadoop-configuration.xml

C hadoop.conf

D hdfs-site.xml

Question: 13
To run a MapReduce job on the BigInsights cluster,
which statement about the input file(s) must be true?
Your answer

A The file(s) must be stored on the local file system where the map reduce job was

B The file(s) must be stored in HDFS or GPFS.

C The file(s) must be stored on the JobTracker.

D No matter where the input files are before, they will be automatically copied to wh
Question: 14
What is a characteristic of IBM GPFS that
distinguishes it from other distributed file systems?
Your answer

A operating system independence

B posix compliance

C no single point of failure

D blocks that are stored on different nodes

Question: 15
Which statement represents a difference between
Pig and Hive?
Your answer

A Pig is used for creating MapReduce programs.

B Pig has a shelf interface for executing commands.

C Pig is not designed for random reads/writes or low-latency queries.

D Pig uses Load, Transform, and Store.

Question: 16

Which command helps you create a directory called

mydata on HDFS?
Your answer

A hdfs -dir mydata

B hadoop fs -mkdir mydata

C hadoop fs -dir mydata

D mkdir mydata

Question: 17

In which step of a MapReduce job is the output

stored on the local disk?
Your answer

A Reduce

B Shuffle

C Combine

D Map

Question: 18
Under the MapReduce programming model, which
task is performed by the Reduce step?
Your answer

A Worker nodes process individual data segments in parallel.

B Worker nodes store results in the local file system.

Your answer

C Input data is split into smaller pieces.

D Data is aggregated by worker nodes.

Question: 19
Which element of the MapReduce architecture runs
map and reduce jobs?
Your answer

A Reducer

B JobScheduler

C TaskTracker

D JobTracker

Question: 20
What is one of the two driving principles of
MapReduce?
Your answer

A spread data across a cluster of computers

B provide structure to unstructured or semi-structured data

C increase storage capacity through advanced compression algorithms

D provide a platform for highly efficient transaction processing

Question: 21
When running a MapReduce job from Eclipse, which
BigInsights execution models are available? (Select
two.)
Your answer

A Cluster

B Distributed

C Remote

D Debugging

E Local

Question: 22
Which statement is true regarding the number of
mappers and reducers configured in a cluster?
Your answer

A The number of reducers is always equal to the number of mappers.

B The number of mappers and reducers can be configured by modifying the mapred-site.xm

C The number of mappers and reducers is decided by the NameNode.

D The number of mappers must be equal to the number of nodes in a cluster.

Question: 23
Which command displays the sizes of files and
directories contained in the given directory, or the
length of a file, in case it is just a file?
Your answer

A hadoop size

B hdfs -du

C hdfs fs size

D hadoop fs -du

Question: 24
Following the most common HDFS replica
placement policy, when the replication factor is
three, how many replicas will be located on the local
rack?
Your answer

A three

B two

C one

D none

Question: 25
In the MapReduce processing model, what is the
main function performed by the JobTracker?
Your answer

A copies Job Resources to the shared file system

B coordinates the job execution

Your answer

C executes the map and reduce functions

D assigns tasks to each cluster node

Question: 26
How are Pig and Jaql query languages similar?
Your answer

A Both are data flow languages.

B Both require schema.

C Both use Jaql query language.

D Both are developed primarily by IBM.

Question: 27
Under the HDFS architecture, what is one purpose of
the NameNode?
Your answer

A to manage storage attached to nodes

B to coordinate MapReduce jobs

C to regulate client access to files

D to periodically report status to DataNode

Question: 28
Which command should be used to list the contents
of the root directory in HDFS?
Your answer

A hadoop fs list

B hdfs root

C hadoop fs -Is /

D hdfs list /

Question: 29
What is one function of the JobTracker in
MapReduce?
Your answer

A runs map and reduce tasks

B keeps the work physically close to the data

C reports status of DataNodes

D manages storage

Question: 30
In addition to the high-level language Pig Latin, what
is a primary component of the Apache Pig platform?
Your answer

A built-in UDFs and indexing

Your answer

B platform-specific SQL libraries

C an RDBMS such as DB2 or MySQL

D runtime environment

Question: 31
Which statement is true about Hadoop Distributed
File System (HDFS)?
Your answer

A Data is accessed through MapReduce.

B Data is designed for random access read/write.

C Data can be processed over long distances without a decrease in performance.

D Data can be created, updated and deleted.

Question: 32
Which is a use-case for Text Analytics?
Your answer

A managing customer information in a CRM database

B sentiment analytics from social media blogs

C product cost analysis from accounting systems

D health insurance cost/benefit analysis from payroll data

Question: 33
Which tool is used to access BigSheets?
Your answer

A BigSheets client

B Microsoft Excel

C Eclipse

D Web Browser

Question: 34
Which technology does Big SQL utilize for access to
shared catalogs?
Your answer

A Hive metastore

B RDBMS

C MapReduce

D HCatalog

Question: 35
Which statement will make an AQL view have
content displayed?
Your answer

A display view <view_name>

B return view <view_name>

Your answer

C output view <view_name>

D export view <view_name>

Question: 36
You work for a hosting company that has data
centers spread across North America. You are trying
to resolve a critical performance problem in which a
large number of web servers are performing far
below expectations. You know that the information
written to log files can help determine the cause of
the problem, but there is too much data to manage
easily. Which type of Big Data analysis is
appropriate for this use case?
Your answer

A Text Analytics

B Stream Computing

C Data Warehousing

D Temporal Analysis

Question: 37
Which utility provides a command-line interface for
Hive?
Your answer

A Thrift client
Your answer

B Hive shell

C Hive SQL client

D Hive Eclipse plugin

Question: 38
What is an accurate description of HBase?
Your answer

A It is a data flow language for structured data based on Ansi-SQL.

B It is a distributed file system that replicates data across a cluster.

C It is an open source implementation of Google's BigTable.

D It is a database schema for unstructured Big Data.

Question: 39
Which Hadoop-related technology provides a user-
friendly interface, which enables business users to
easily analyze Big Data?
Your answer

A BigSQL

B BigSheets

C Avro

D HBase
Question: 40
What drives the demand for Text Analytics?
Your answer

A Text Analytics is the most common way to derive value from Big Data.

B MapReduce is unable to process unstructured text.

C Data warehouses contain potentially valuable information.

D Most of the world's data is in unstructured or semi-structured text.

Question: 41
In Hive, what is the difference between an external
table and a Hive managed table?
Your answer

A An external table refers an existing location outside the warehouse directory.

B An external table refers to a table that cannot be dropped.

C An external table refers to the data from a remote database.

D An external table refers to the data stored on the local file system.

Question: 42
Which statement about NoSQL is true?
Your answer

A It provides all the capabilities of an RDBMS plus the ability to manage Big Data.

B It is a database technology that does not use the traditional relational model.
Your answer

C It is based on the highly scalable Google Compute Engine.

D It is an IBM project designed to enable DB2 to manage Big Data.

Question: 43
If you need to JOIN data from two workbooks, which
operation should be performed beforehand?
Your answer

A "Copy" to create a new sheet with the other workbook data in the current workboo

B "Group" to bring together the two workbooks

C "Load" to create a new sheet with the other workbook data in the current workboo

D "Add" to add the other workbook data to the current workbook

Question: 44
What is the "scan" command used for in HBase?
Your answer

A to get detailed information about the table

B to view data in an Hbase table

C to report any inconsistencies in the database

D to list all tables in Hbase

Question: 45

Which tool is used for developing a BigInsights Text

Analytics extractor?
Your answer

A Eclipse with BigInsights tools for Eclipse plugin

B BigInsights Console with AQL plugin

C AQLBuilder

D AQL command line

Question: 46
What is the most efficient way to load 700MB of data
when you create a new HBase table?
Your answer

A Pre-create regions by specifying splits in create table command and use the inser

B Pre-create regions by specifying splits in create table command and bulk loading

C Pre-create the column families when creating the table and use the put command

D Pre-create the column families when creating the table and bulk loading the data.

Question: 47
The following sequence of commands is executed:
create 'table_1','column_family1','column_family2'
put 'table_1','row1','column_family1:c11','r1v11'
put 'table_1','row2','column_family1:c12','r1v12'
put 'table_1','row2','column_family2:c21','r1v21'
put 'table_1','row3','column_family1:d11','r1v11'
put 'table_1','row2','column_family1:d12','r1v12'
put 'table_1','row2','column_family2:d21','r1v21'
In HBase, which value will the "count 'table_1'"
command return?
Your answer

A 4

B 3

C 6

D 2

Question: 48
Which Hive command is used to query a table?
Your answer

A TRANSFORM

B SELECT

C GET

D EXPAND

Question: 49
Why develop SQL-based query languages that can
access Hadoop data sets?
Your answer

A because SQL enhances query performance

B because the MapReduce Java API is sometimes difficult to use

C because data stored in a Hadoop cluster lends itself to structured SQL queries

D because the data stored in Hadoop is always structured

Question: 50
Which key benefit does NoSQL provide?
Your answer

A It allows Hadoop to apply the schema-on-ingest model to unstructured Big Data.

B It allows an RDBMS to maintain referential integrity on a Hadoop data set.

C It allows customers to leverage high-end server platforms to manage Big Data.

D It can cost-effectively manage data sets too large for traditional RDBMS.

Question: 51
What makes SQL access to Hadoop data difficult?
Your answer

A Hadoop data is highly structured.

B Data is in many formats.

C Data is located on a distributed file system.

D Hadoop requires pre-defined schema.

Question: 52
Which command can be used in Hive to list the
tables available in a database/schema?
Your answer

A list tables

B describe tables

C show all

D show tables

Question: 53
In HBase, what is the "count" command used for?
Your answer

A to count the number of columns of a table

B to count the number of column families of a table

C to count the number of rows in a table

D to count the number of regions of a table

Question: 54
Which Hadoop-related technology supports analysis
of large datasets stored in HDFS using an SQL-like
query language?
Your answer

A HBase
Your answer

B Pig

C Jaql

D Hive

Question: 55
How can the applications published to BigInsights
Web Console be made available for users to
execute?
Your answer

A They need to be marked as "Shared."

B They need to be copied under the user home directory.

C They need to be deployed with proper privileges.

D They need to be linked with the master application.

Question: 56
Which component of Apache Hadoop is used for
scheduling and running workflow jobs?
Your answer

A Eclipse

B Oozie

C Jaql

D Task Launcher
Question: 57
What is one of the main components of Watson
Explorer (InfoSphere Data Explorer)?
Your answer

A validater

B replicater

C crawler

D compressor

Question: 58
IBM InfoSphere Streams is designed to accomplish
which Big Data function?
Your answer

A analyze and react to data in motion before it is stored

B find and analyze historical stream data stored on disk

C analyze and summarize product sentiments posted to social media

D execute ad-hoc queries against a Hadoop-based data warehouse

Question: 59
Which IBM Big Data solution provides low-latency
analytics for processing data-in-motion?
Your answer

A InfoSphere Information Server

Your answer

B InfoSphere Streams

C InfoSphere BigInsights

D PureData for Analytics

Question: 60
Which IBM tool enables BigInsights users to
develop, test and publish BigInsights applications?
Your answer

A Avro

B HBase

C Eclipse

D BigInsights Applications Catalog

Question: 5
Which description identifies the real value of Big Data and An
Your answer

A enabling customers to efficiently index and access large volumes of data

B gaining new insight through the capabilities of the world's interconnected intellige

C providing solutions to help customers manage and grow large database systems

D using modern technology to efficiently store the massive amounts of data generat
 examen blanc big data
question reponses
Question: 1 A Jaql
Which capability does IBM BigInsights add to enrich Hadoop? B Fault tolerance through HDFS replication
C Adaptive MapReduce
D Parallel computing on commodity servers

Question: 2 A value
What is one of the four characteristics of Big Data? B volume
C verifiability
D volatility

Question: 3 A Hadoop Common

Which Hadoop-related project provides common utilities and libraries that support other B Hadoop HBase
Hadoop sub projects? C MapReduce
D BigTable
Question: 4 Which type of Big Data analysis involves A Federated Discovery and Navigation
the processing of extremely large volumes of constantly moving data that is impractical to B Text Analysis
store? C Stream Computing
D MapReduce

Question: 6 A 64-bit architecture

Which primary computing bottleneck of modern computers is addressed by Hadoop? B disk latency
C MIPS
D limited disk capacity
Question: 7 A stream computing
Which Big Data function improves the decision-making capabilities of organizations by B data warehousing
enabling the organizations to interpret and evaluate structured and unstructured data in C analytics
search of valuable business information? D distributed file system

Question: 8 A HBase
What is one of the two technologies that Hadoop uses as its foundation? B Apache
C Jaql
D MapReduce
Question: 9 A a high throughput, shared file system
What key feature does HDFS 2.0 provide that HDFS does not? B high availability of the NameNode
C data access performed by an RDBMS
D random access to data in the cluster
Question: 10 A LOAD
What are two of the core operators that can be used in a Jaql query? (Select two.) B JOIN
C TOP
D SELECT
Question: 11 A SQL-like
Which type of language is Pig? B compiled language
C object oriented
D data flow
If you need to change the replication factor or increase the default storage block size, which A hdfs.conf
file do you need to modify? B hadoop-configuration.xml
C hadoop.conf
D hdfs-site.xml
Question: 13 A The file(s) must be stored on the local file system where the map reduce job was
To run a MapReduce job on the BigInsights cluster, which statement about the input file(s) developed.
must be true? B The file(s) must be stored in HDFS or GPFS.
C The file(s) must be stored on the JobTracker.
D No matter where the input files are before, they will be automatically copied to where
the job runs.
What is a characteristic of IBM GPFS that distinguishes it from other distributed file systems? A operating system independence
B posix compliance
C no single point of failure
D blocks that are stored on different nodes
Question: 15 A Pig is used for creating MapReduce programs.
Which statement represents a difference between Pig and Hive? B Pig has a shelf interface for executing commands.
C Pig is not designed for random reads/writes or low-latency queries.
D Pig uses Load, Transform, and Store.

Question: 16 A hdfs -dir mydata

B hadoop fs -mkdir mydata
Which command helps you create a directory called mydata on HDFS? C hadoop fs -dir mydata
D mkdir mydata

Question: 17 A Reduce
B Shuffle
In which step of a MapReduce job is the output stored on the local disk? C Combine
D Map

Question: 18 A Worker nodes process individual data segments in parallel.

Under the MapReduce programming model, which task is performed by the Reduce step? B Worker nodes store results in the local file system.
C Input data is split into smaller pieces.
D Data is aggregated by worker nodes.

Question: 19 A Reducer
Which element of the MapReduce architecture runs map and reduce jobs? B JobScheduler
C TaskTracker
D JobTracker

Question: 19 A Reducer
Which element of the MapReduce architecture runs map and reduce jobs? B JobScheduler
C TaskTracker
D JobTracker

Question: 20 A spread data across a cluster of computers

What is one of the two driving principles of MapReduce? B provide structure to unstructured or semi-structured data
C increase storage capacity through advanced compression algorithms
D provide a platform for highly efficient transaction processing

Question: 21 A Cluster
When running a MapReduce job from Eclipse, which BigInsights execution models are B Distributed
available? (Select two.) C Remote
D Debugging
E Local
Question: 22 A The number of reducers is always equal to the number of mappers.
Which statement is true regarding the number of mappers and reducers configured in a B The number of mappers and reducers can be configured by modifying the mapred-site.
cluster? xml file.
C The number of mappers and reducers is decided by the NameNode.
D The number of mappers must be equal to the number of nodes in a cluster.

Question: 23 A hadoop size

Which command displays the sizes of files and directories contained in the given directory, or B hdfs -du
the length of a file, in case it is just a file? C hdfs fs size
D hadoop fs -du

Question: 24 A three
Following the most common HDFS replica placement policy, when the replication factor is B two
three, how many replicas will be located on the local rack? C one
D none

Question: 25 A copies Job Resources to the shared file system

In the MapReduce processing model, what is the main function performed by the JobTracker? B coordinates the job execution
C executes the map and reduce functions
D assigns tasks to each cluster node

Question: 26 A Both are data flow languages.

How are Pig and Jaql query languages similar? B Both require schema.
C Both use Jaql query language.
D Both are developed primarily by IBM.

Question: 27 A to manage storage attached to nodes

Under the HDFS architecture, what is one purpose of the NameNode? B to coordinate MapReduce jobs
C to regulate client access to files
D to periodically report status to DataNode

Question: 28 A hadoop fs list

Which command should be used to list the contents of the root directory in HDFS? B hdfs root
C hadoop fs -Is /
D hdfs list /

Question: 29 A runs map and reduce tasks

What is one function of the JobTracker in MapReduce? B keeps the work physically close to the data
C reports status of DataNodes
D manages storage

Question: 30 A built-in UDFs and indexing

In addition to the high-level language Pig Latin, what is a primary component of the Apache B platform-specific SQL libraries
Pig platform? C an RDBMS such as DB2 or MySQL
D runtime environment

Question: 31 A Data is accessed through MapReduce.

Which statement is true about Hadoop Distributed File System (HDFS)? B Data is designed for random access read/write.
C Data can be processed over long distances without a decrease in performance.
D Data can be created, updated and deleted.
Question: 32 A managing customer information in a CRM database
Which is a use-case for Text Analytics? B sentiment analytics from social media blogs
C product cost analysis from accounting systems
D health insurance cost/benefit analysis from payroll data

Question: 33 A BigSheets client

Which tool is used to access BigSheets? B Microsoft Excel
C Eclipse
D Web Browser

Question: 34 A Hive metastore

Which technology does Big SQL utilize for access to shared catalogs? B RDBMS
C MapReduce
D HCatalog

Question: 35 A display view <view_name>

Which statement will make an AQL view have content displayed? B return view <view_name>
C output view <view_name>
D export view <view_name>

Question: 36 A Text Analytics

You work for a hosting company that has data centers spread across North America. You are B Stream Computing
trying to resolve a critical performance problem in which a large number of web servers are C Data Warehousing
performing far below expectations. You know that the information written to log files can help D Temporal Analysis
determine the cause of the problem, but there is too much data to manage easily. Which type
of Big Data analysis is appropriate for this use case?

Question: 37 A Thrift client

Which utility provides a command-line interface for Hive? B Hive shell
C Hive SQL client
D Hive Eclipse plugin

Question: 38 A It is a data flow language for structured data based on Ansi-SQL.

What is an accurate description of HBase? B It is a distributed file system that replicates data across a cluster.
C It is an open source implementation of Google's BigTable.
D It is a database schema for unstructured Big Data.

Question: 39 A BigSQL
Which Hadoop-related technology provides a user-friendly interface, which enables business B BigSheets
users to easily analyze Big Data? C Avro
D HBase

Question: 40 A Text Analytics is the most common way to derive value from Big Data.
What drives the demand for Text Analytics? B MapReduce is unable to process unstructured text.
C Data warehouses contain potentially valuable information.
D Most of the world's data is in unstructured or semi-structured text.

Question: 41 A An external table refers an existing location outside the warehouse directory.
In Hive, what is the difference between an external table and a Hive managed table? B An external table refers to a table that cannot be dropped.
C An external table refers to the data from a remote database.
D An external table refers to the data stored on the local file system.
Question: 42 A It provides all the capabilities of an RDBMS plus the ability to manage Big Data.
Which statement about NoSQL is true? B It is a database technology that does not use the traditional relational model.
C It is based on the highly scalable Google Compute Engine.
D It is an IBM project designed to enable DB2 to manage Big Data.

Question: 43 A "Copy" to create a new sheet with the other workbook data in the current workbook
If you need to JOIN data from two workbooks, which operation should be performed B "Group" to bring together the two workbooks
beforehand? C "Load" to create a new sheet with the other workbook data in the current workbook
D "Add" to add the other workbook data to the current workbook

Question: 44 A to get detailed information about the table

What is the "scan" command used for in HBase? B to view data in an Hbase table
C to report any inconsistencies in the database
D to list all tables in Hbase

Which tool is used for developing a BigInsights Text Analytics extractor? A Eclipse with BigInsights tools for Eclipse plugin
B BigInsights Console with AQL plugin
C AQLBuilder
D AQL command line

Question: 46 A Pre-create regions by specifying splits in create table command and use the insert
What is the most efficient way to load 700MB of data when you create a new HBase table? command to load data.
B Pre-create regions by specifying splits in create table command and bulk loading the
data.
C Pre-create the column families when creating the table and use the put command to
load the data.
D Pre-create the column families when creating the table and bulk loading the data.

Question: 47 A4
The following sequence of commands is executed: B3
create 'table_1','column_family1','column_family2' C6
put 'table_1','row1','column_family1:c11','r1v11' D2
put 'table_1','row2','column_family1:c12','r1v12'
put 'table_1','row2','column_family2:c21','r1v21'
put 'table_1','row3','column_family1:d11','r1v11'
put 'table_1','row2','column_family1:d12','r1v12'
put 'table_1','row2','column_family2:d21','r1v21'
In HBase, which value will the "count 'table_1'" command return?

Question: 48 A TRANSFORM
Which Hive command is used to query a table? B SELECT
C GET
D EXPAND

Question: 49 A because SQL enhances query performance

Why develop SQL-based query languages that can access Hadoop data sets? B because the MapReduce Java API is sometimes difficult to use
C because data stored in a Hadoop cluster lends itself to structured SQL queries
D because the data stored in Hadoop is always structured

Question: 50 A It allows Hadoop to apply the schema-on-ingest model to unstructured Big Data.
Which key benefit does NoSQL provide? B It allows an RDBMS to maintain referential integrity on a Hadoop data set.
C It allows customers to leverage high-end server platforms to manage Big Data.
D It can cost-effectively manage data sets too large for traditional RDBMS.
Question: 51 A Hadoop data is highly structured.
What makes SQL access to Hadoop data difficult? B Data is in many formats.
C Data is located on a distributed file system.
D Hadoop requires pre-defined schema.

Question: 52 A list tables

Which command can be used in Hive to list the tables available in a database/schema? B describe tables
C show all
D show tables

Question: 53 A to count the number of columns of a table

In HBase, what is the "count" command used for? B to count the number of column families of a table
C to count the number of rows in a table
D to count the number of regions of a table

Question: 54 A HBase
Which Hadoop-related technology supports analysis of large datasets stored in HDFS using an B Pig
SQL-like query language? C Jaql
D Hive

Question: 55 A They need to be marked as "Shared."

How can the applications published to BigInsights Web Console be made available for users to B They need to be copied under the user home directory.
execute? C They need to be deployed with proper privileges.
D They need to be linked with the master application.

Question: 56 A Eclipse
Which component of Apache Hadoop is used for scheduling and running workflow jobs? B Oozie
C Jaql
D Task Launcher
Question: 57 A validater
What is one of the main components of Watson Explorer (InfoSphere Data Explorer)? B replicater
C crawler
D compressor

Question: 58 A analyze and react to data in motion before it is stored

IBM InfoSphere Streams is designed to accomplish which Big Data function? B find and analyze historical stream data stored on disk
C analyze and summarize product sentiments posted to social media
D execute ad-hoc queries against a Hadoop-based data warehouse

Question: 59 A InfoSphere Information Server

Which IBM Big Data solution provides low-latency analytics for processing data-in-motion? B InfoSphere Streams
C InfoSphere BigInsights
D PureData for Analytics

Question: 60 A Avro
Which IBM tool enables BigInsights users to develop, test and publish BigInsights applications? B HBase
C Eclipse
D BigInsights Applications Catalog
Question: 5 A enabling customers to efficiently index and access large volumes of data
Which description identifies the real value of Big Data and Analytics? B gaining new insight through the capabilities of the world's interconnected intelligence
C providing solutions to help customers manage and grow large database systems
D using modern technology to efficiently store the massive amounts of data generated
by social networks

examen big data

question reponses
Question: 1 A Notebook
Where must a Spark configuration be set up first? B Db2 Warehouse
C IBM Cloud
D Watson Data Platform

Question: 2 A Watson Studio homepage

When sharing a notebook, what will always point to the most recent version of the notebook? B The permalink
C The Spark service
D PixieDust visualization

Question: 3 A Spark service

When creating a Watson Studio project, what do you need to specify? B Data service
C Collaborators
D Data assets

Question: 4 AR
You can import preinstalled libraries if you are using which languages? (Select two.) B Python
(Please select ALL that apply) C Bash
D Rexx
E Scala

Question: 5 A Viewers
Who can control a Watson Studio project assets? B Editors
C Collaborators
D Tenants

Question: 6 A ZOOKEEPER_APP
Which environmental variable needs to be set to properly start ZooKeeper? B ZOOKEEPER_DATA
C ZOOKEEPER
D ZOOKEEPER_HOME

Question: 7 A better compression using GZip

Which is the primary advantage of using column-based data formats over record-based B supports in-memory processing
formats? C facilitates SQL-based queries
D faster query execution

Question: 8 A Collect and send data into a stream.

What is the primary purpose of Apache NiFi? B Finding data across the cluster.
C Connect remote data sources via WiFi.
D Identifying non-compliant data access.
Question: 9 A cash register receipts
What are three examples of Big Data? (Choose three.) B web server logs
(Please select ALL that apply) C inventory database records
D bank records
E photos posted on Instragram
F messages tweeted on Twitter

Question: 10 A get /
What ZK CLI command is used to list all the ZNodes at the top level of the ZooKeeper B create /
hierarchy, in the ZooKeeper command-line interface? C listquota /
D ls /

Question: 11 A JSON
What is the default data format Sqoop parses to export data to a database? B CSV
C XML
D SQL

Question: 12 A Keeps the tasks physically close to the data.

Under the MapReduce v1 architecture, which function is performed by the TaskTracker? B Pushes map and reduce tasks out to DataNodes.
C Manages storage and transmission of intermediate output.
D Accepts MapReduce jobs submitted by clients.

Question: 13 A Indexed databases containing very large volumes of historical data used for compliance
Which statement describes "Big Data" as it is used in the modern business world? reporting purposes.
B Non-conventional methods used by businesses and organizations to capture, manage,
process, and make sense of a large volume of data.
C Structured data stores containing very large data sets such as video and audio streams.
D The summarization of large indexed data stores to provide information about potential
problems or opportunities.

Question: 14 A Runs map and reduce tasks.

Under the MapReduce v1 architecture, which function is performed by the JobTracker? B Accepts MapReduce jobs submitted by clients.
C Manages storage and transmission of intermediate output.
D Reports status to MasterNode.

Question: 15 A HDFS is a software framework to support computing on large clusters of computers.

Which statement is true about the Hadoop Distributed File System (HDFS)? B HDFS provides a web-based tool for managing Hadoop clusters.
C HDFS links the disks on multiple nodes into one large file system.
D HDFS is the framework for job scheduling and cluster resource management.

Question: 16 A Coordination between servers.

How does MapReduce use ZooKeeper? B Aid in the high availability of Resource Manager.
C Master server election and discovery.
D Server lease management of nodes.

Question: 17 A Python
Which two Spark libraries provide a native shell? (Choose two.) B Scala
(Please select ALL that apply) C C#
D Java
E C++
Question: 18 A IP address
What is an authentication mechanism in Hortonworks Data Platform? B Preshared keys
C Kerberos
D Hardware token

Question: 19 A Manage, secure, and govern data stored across all storage environments.
What is Hortonworks DataPlane Services (DPS) used for? B Transform data from CSV format into native HDFS data.
C Perform backup and recovery of data in the Hadoop ecosystem.
D Keep data up to date by periodically refreshing stale data.

Question: 20 A Copy any appropriate JDBC driver JAR to $SQOOP_HOME/lib.

What must be done before using Sqoop to import from a relational database? B Complete the installation of Apache Accumulo.
C Create a Java class to support the data import.
D Create an empty database for Sqoop to access.

Question: 21 A Scala
What is the native programming language for Spark? B C++
C Java
D Python

Question: 22 A YARN
Which Hortonworks Data Platform (HDP) component provides a common web user interface B HDFS
for applications running on a Hadoop cluster? C Ambari
D MapReduce
Question: 23 A Transformations
Which Spark RDD operation returns values after performing the evaluations? B Actions
C Caching
D Evaluations

Question: 24 A Configuration bootstrapping for new nodes.

Which two are use cases for deploying ZooKeeper? (Choose two.) B Managing the hardware of cluster nodes.
(Please select ALL that apply) C Storing local temporary data files.
D Simple data registry between nodes.

Question: 25 A DataNodes increase capacity while NameNodes increase processing power.

In a Hadoop cluster, which two are the result of adding more nodes to the cluster? (Choose B It adds capacity to the file system.
two.) C Scalability increases by a factor of x^N-1.
(Please select ALL that apply) D Capacity increases while fault tolerance decreases.
E It increases available processing power.

Question: 26 A Distribution
Which Spark RDD operation creates a directed acyclic graph through lazy evaluations? B GraphX
C Transformations
D Actions

Question: 27 A REST APIs

Which feature allows application developers to easily use the Ambari interface to integrate B Postgres RDBMS
Hadoop provisioning, management, and monitoring capabilities into their own applications? C Ambari Alert Framework
D AMS APIs
Question: 28 A Columns of data must be separated by a delimiter.
What is one disadvantage to using CSV formatted data in a Hadoop data store? B Fields must be positioned at a fixed offset from the beginning of the record.
C It is difficult to represent complex data structures such as maps.
D Data must be extracted, cleansed, and loaded into the data warehouse.

Question: 29 A YARN
Which element of Hadoop is responsible for spreading data across the cluster? B MapReduce
C AMS
D HDFS

Question: 30 A Authorization Provider

Which component of the Apache Ambari architecture stores the cluster configurations? B Ambari Metrics System
C Postgres RDBMS
D Ambari Alert Framework

Question: 31 A Time of interaction

Which two are examples of personally identifiable information (PII)? (Choose two.) B Medical record number
(Please select ALL that apply) C Email address
D IP address

Question: 32 A SlaveNode
Under the MapReduce v1 architecture, which element of the system manages the map and B JobTracker
reduce functions? C MasterNode
D StorageNode
E TaskTracker

Question: 33 A NameNode
Which component of the HDFS architecture manages storage attached to the nodes? B StorageNode
C DataNode
D MasterNode

Question: 34 A Volume
Which of the "Five V's" of Big Data describes the real purpose of deriving business insight from B Value
Big Data? C Variety
D Velocity
E Veracity

Question: 35 A Spark Learning

Which component of the Spark Unified Stack supports learning algorithms such as, logistic B Spork
regression, naive Bayes classification, and SVM? C Spark SQL
D MLlib

Question: 36 A able to use inexpensive commodity hardware

Which two descriptions are advantages of Hadoop? (Choose two.) B intensive calculations on small amounts of data
(Please select ALL that apply) C processing a large number of small files
D processing random access transactions
E processing large volumes of data with high throughput

Question: 37 A CSV
Which two of the following are row-based data encoding formats? (Choose two.) B Avro
(Please select ALL that apply) C ETL
D Parquet
E RC and ORC
Question: 38 A The data is spread out and replicated across the cluster.
Which statement describes the action performed by HDFS when data is written to the Hadoop B The data is replicated to at least 5 different computers.
cluster? C The MasterNodes write the data to disk.
D The FsImage is updated with the new data map.

Question: 39 A MasterNode
Under the MapReduce v1 architecture, which element of MapReduce controls job execution B JobTracker
on multiple slaves? C SlaveNode
D TaskTracker
E StorageNode
Question: 40 A MLlib
Which component of the Spark Unified Stack provides processing of data arriving at the B Spark SQL
system in real-time? C Spark Streaming
D Spark Live

Question: 41 A DB2ATSENABLE
Which two registries are used for compiler and runtime performance improvements in B DB2F ODC
support of the Big SQL environment? (Choose two) C DB2COMPOPT
(Please select ALL that apply) D DB2RSHTIMEOUT
E DB2SORTAFTER_TQ

Question: 42 A bigsql_bar.py
Which script is used to backup and restore the Big SQL database? B db2.sh
C bigsql.sh
D load.py

Question: 43 A STRING
You need to create a table that is not managed by the Big SQL database manager. Which B BOOLEAN
keyword would you use to create the table? C SMALLINT
D EXTERNAL

Question: 44 A Oracle
Which two of the following data sources are currently supported by Big SQL? (Choose two) B PostgreSQL
(Please select ALL that apply) C Teradata
D MySQL
E MariaDB

Question: 45 A 7055
Which port is the default for the Big SQL Scheduler to get administrator commands? B 7054
C 7052
D 7053

Question: 46 A Hortonworks
Which tool should you use to enable Kerberos security? B Ambari
C Apache Ranger
D Hive

Question: 47 A Scheduler
Which two options can be used to start and stop Big SQL? (Choose two) B DSM Console
(Please select ALL that apply) C Command line
D Java SQL shell
Question: 48 A CREATE
Which command is used to populate a Big SQL table? B QUERY
C SET
D LOAD

Question: 49 A Impersonation
Which feature allows the bigsql user to securely access data in Hadoop on behalf of another B Privilege
user? C Rights
D Schema

Question: 50 A SET AUTHORIZATION

Which command would you run to make a remote table accessible using an alias? B CREATE SERVER
C CREATE WRAPPER
D CREATE NICKNAME

Question: 51 A Db2
The Big SQL head node has a set of processes running. What is the name of the service ID B hdfs
running these processes? C user1
D bigsql

Question: 52 A Parquet
Which file format contains human-readable data where the column values are separated by a B ORC
comma? C Delimited
D Sequence

Question: 53 A Public key

Which Big SQL authentication mode is designed to provide strong authentication for B Flat files
client/server applications by using secret-key cryptography? C Kerberos
D LDAP

Question: 54 A Jupyter
Which type of foundation does Big SQL build on? B Apache HIVE
C RStudio
D MapReduce

Question: 55 A SSL
You need to monitor and manage data security across a Hadoop platform. Which tool would B HDFS
you use? C Hive
D Apache Ranger

Question: 56 A """
What can be used to surround a multi-line string in a Python code cell by appearing before and B "
after the multi-line string? C

Question: 57 A Packaging data for public distribution on a website.

For what are interactive notebooks used by data scientists? B Quick data exploration tasks that can be reproduced.
C Providing a chain of custody of all data.
D Bulk loading data into a database.

Question: 58 A pull
What Python statement is used to add a library to the current code cell? B import
C load
D using
Question: 59 A NLTK
What Python package has support for linear algebra, optimization, mathematical integration, B Pandas
and statistics? C NumPy
D SciPy

Question: 60 A Traditional research

Which three main areas make up Data Science according to Drew Conway? (Choose three.) B Machine learning
(Please select ALL that apply) C Substantive expertise
D Math and statistics knowledge
E Hacking skills

Big Data Engineer v2IBM Certification2018

question reponses
1/ What are the 4Vs of Big Data? (Please select the FOUR that apply) •Veracity •Velocity •Variety •Volume
2/ What are the three types of Big Data? (Please select the THREE that apply) •Semi-structured •Structured •Unstructured
3/ Select all the components of HDP which provides data access capabilities •Pig •MapReduce •Hive
4/ Select the components that provides the capability to move data from relational database
•Sqoop •Kafka •Flume
into Hadoop.
5/ Managing Hadoop clusters can be accomplished using which component? Ambari
6/ True or False: The following components are value-add from IBM: Big Replicate, Big SQL,
TRUE
BigIntegrate, BigQuality, Big Match
7/ True or False: Data Science capabilities can be achieved using only HDP. FALSE(Big Data Ecosystem UNIT 2)
8/ True or False: Ambari is backed by RESTful APIs for developers to easily integrate with their
True
own applications.
9/ Which Hadoop functionalities does Ambari provide? Manage •Provision •Integrate •Monitor
10/ Which page from the Ambari UI allows you to check the versions of the software installed
The Admin > Manage Ambari page
on your cluster?
11/ True or False?Creating users through the Ambari UI will also create the user on the HDFS. FALSE
12/ True or False? You can use the CURL commands to issue commands to Ambari TRUE
13/ True or False: Hadoop systems are designed for transaction processing. FALSE
14/ What is the default number of replicas in a Hadoop system? 3
15/ True or False: One of the driving principal of Hadoop is that the data is brought to the
FALSE(Big Data Ecosystem UNIT 4)
program.
16/ True or False: Atleast 2 Name Nodes are required for a standalone Hadoop cluster. FALSE(Big Data Ecosystem UNIT 4)
17/ True or False: The phases in a MR job are Map, Shuffle, Reduce and Combiner TRUE
18/ Centralized handling of job control flow is one of the the limitations of MRv1 TRUE
19/ The Job Tracker in MR1 is replaced by which component(s) in YARN? •ApplicationMaster •ResourceManager
20/ What are the benefits of using Spark? (Please select the THREE that apply) •Generality •Speed •East of use
21/ What are the languages supported by Spark? (Please select the THREE that apply) •Python •Java •Scala
22/ Resilient Distributed Dataset (RDD) is the primary abstraction of Spark. True
23/ What would you need to do in a Spark application that you would not need to do in a
Import the necessary libraries to load the SparkContext
Spark shell to start using Spark?
24/ True or False: NoSQL database is designed for those that do not want to use SQL. FALSE(Big Data Ecosystem UNIT 7
25/ Which database is a columnar storage database? Hbase
26/ Which database provides a SQL for Hadoop interface? Hive
27/ Which Apache project provides coordination of resources? ZooKeeper
28/ What is ZooKeeper's role in the Hadoop infrastructure? •Manage the coordination between HBase servers •Hadoop and MapReduce uses
ZooKeeper to aid in high availability of Resource Manager •Flume uses
ZooKeeper for configuration purposes in recent releases
29/ True or False: Slider provides an intuitive UI which allows you to dynamically allocate YARN
FALSE(Big Data Ecosystem UNIT 8)
resources.
30/ True or False: Knox can provide all the security you need within your Hadoop
FALSE(Big Data Ecosystem UNIT 8)
infrastructure.
31/ True or False: Sqoop is used to transfer data between Hadoop and relational databases. True
32/True or False: For Sqoop to connect to a relational database, the JDBC JAR files for that
FALSE(Big Data Ecosystem UNIT 9)
database must be located in $SQOOP_HOME/bin.
33/ True or False: Each Flume node receives data as "source", stores it in a "channel", and
True
sends it via a "sink".
34/ Through what HDP component are Kerberos, Knox, and Ranger managed? Ambari
35/ Which security component is used to provide peripheral security? Apache Knox
36/ One of the governance issue that Hortonworks DataPlane Service (DPS) address is visibility
over all of an organization's data across all of their environments — on-prem, cloud, hybrid — True
while making it easy to maintain consistent security and governance
37/ True or false: The typical sources of streaming data are Sensors, "Data exhaust" and high-
True
rate transaction data.
38/ What are the components of Hortonworks Data Flow(HDF)? •Flow management •Stream processing •Enterprise services
39/ True or False: NiFi is a disk-based, microbatch ETL tool that provides flow management True
40/ True or False: MiNiFi is a complementary data collection tool that feeds collected data to
True
NiFi
41/ What main features does IBM Streams provide as a Streaming Data Platform? (Please
•Analysis and visualization •Rich data connections •Development support
select the THREE that apply)
42/ What are the most important computer languages for Data Analytics?(Please select the
•Python •R •Scala
THREE that apply
43/ True or False: GPUs are special-purpose processors that traditionally can be used to power
graphical displays, but for Data Analytics lend themselves tofaster algorithm execution True
because of the large number of independent processing cores.
44/ True or False: Jupyter stores its workbooks in files with the .ipynb suffix. These files can
FALSE(Introduction to Data Science UNIT 1)
not be stored locally or on a hub server.
45/ $BIGSQL_HOME/bin/bigsql startcommand is used to start Big SQL from the command
True
line?
46/ What are the two ways you can work with Big SQL.(Please select the TWO that apply) •Jsqsh •Web tooling from DSM
47/ What is one of the reasons to use Big SQL? Want to access your Hadoop data without using MapReduce
48/ Should you use the default STRING data type? No(Big SQL UNIT 2)
49/The BOOLEAN type is defined as SMALLINT SQL type in Big SQL. True
50/ Using the LOAD operation is the recommended method for getting data into your Big SQL
True
table for best performance.
51/ Which file storage format has the highest performance? Parquet
52/ What are the two ways to classify functions? •Built-in functions •User-defined functions
53/ True or False: UMASK is used to determine permissions on directories and files. True
54/ True or False: You can only Kerberize a Big SQL server before it is installed. False(Big SQL UNIT 4)
55/ True or False: Authentication with Big SQL only occurs at the Big SQL layer or the client's
False(Big SQL UNIT 4)
application layer.
56/ True or False: Ranger and impersonation works well together. False(Big SQL UNIT 4)
57/ True or False: RCAC can hide rows and columns. True
58/ True or False: Nicknames can be used for wrappers and servers. False(Big SQL UNIT 5)
59/ True or False: Server objects defines the property and values of the connection. True
60/ True or False: The purpose of a wrapper provide a library of routines that doesn't
False(Big SQL UNIT 5)
communicates with the data source.
61/ True or False: User mappings are used to authenticate to the remote datasource. True
62/ True or False: Collaboration with Watson Studio is an optional add-on component that
False(Watson Studio UNIT 1)
must be purchased.
63/ True or False: Watson Studio is designed only for Data Scientists, other personas would
False(Watson Studio UNIT 1)
not know how to use it.
64/ True or False: Community provides access to articles, tutorials, and even data sets that you
True
can use.
65/ True or False: You can import visualization libraries into Watson Studio. True
66/ True or False: Collaborators can be given certain access levels. True
67/True or False: Watson Studio contains Zeppelin as a notebook interface. False(Watson Studio UNIT 2) Jupyter as notebook
Big Data QCM
question reponses
Which component connects sinks and sources in Flume? A. HDFS
B. ElasticSearch
* C. channels
D. Interceptors
Why does YARN scale better than Hadoop v1 for multiple jobs? (Choose two.) A. There is one Job Tracker per cluster.
(Please select ALL that apply) * B. Job tracking and resource management are split.
C. Job tracking and resource management are one process.
* D. There is one Application Master per job.
What happens if a task fails during a Hadoop job execution? A. The job will be restarted with different compute nodes.
B. The entire job will fail.
C. The job will finish with incomplete results.
* D. The task will be restarted on another node.
What command will list files located on the HDFS in R? A. bigr.dir()
B. ls()
* C. bigr.listfs()
D. list()

Which Big SQL datatype should be avoided because it causes significant performance A. CHAR
degradation? * B. STRING
C. UNION
D. VARCHAR
You need to create multiple Big SQL tables with columns defined as CHAR. What needs to be * A. SET SYSHADOOP.COMPATIBILITY_MODE=1
set to enable CHAR columns? B. CREATE TABLE chartab
C. SET HADOOPCOMPATIBLITY_MODE=True
D. ALTER CHAR DATATYPE TO byte
What is the primary core abstraction of Apache Spark? A. GraphX
* B. Resilient Distributed Dataset (RDD)
C. Spark Streaming
D. Directed Acyclic Graph (DAG)

Which Text Analytics runtime component is used for languages such as Spanish and English by A. Named entity extractors
breaking a stream of text into phrases or words? B. Other extractors
* C. Standard tokenizer
D. Multilingual tokenizer
Question : Which two commands are used to load data into an existing Big SQL table from * A. Load
HDFS? (Choose two.) B. Table
(Please select ALL that apply) C. Select
* D. Insert
E. Create
Which command should you use to set the default schema in a Big SQL table and also create A. default
the schema if it does not exist? B. create
C. format
* D. use
What is missing from the following statement when querying a remote table? CREATE A. TABLE
_______ FOR remotetable1 … B. VIEW
* C. NICKNAME
D. INDEX

What are two major business advantages of using BigSheets? (Choose two.) * A. built-in data readers for multiple formats
(Please select ALL that apply) * B. spreadsheet-like querying and discovery interface
C. command-line-driven data analysis
D. feature rich programming environment

Where should you build extractors in the Information Extraction Web Tool? A. Documents
* B. Canvas
C. Property pane
D. Regular expression

In which text analytics phase are extractors developed and tested? A. Analysis
* B. Rule Development
C. Production
D. Performance Tuning
Which action is performed during the Reduce step of a MapReduce v1 processing cycle? * A. Intermediate results are aggregated.
B. The TaskTracker distributes the job to the cluster.
C. The initial problem is broken into pieces.
D. The JobTrackers execute their assigned tasks.

What are two benefits of using the IBM Big SQL processing engine? (Choose two.) A. Core functionality is written in Java for portability.
(Please select ALL that apply) B. The system is built to be started and stopped on demand.
* C. Various data storage formats are supported.
* D. It provides access to Hadoop data using SQL.
An organization is developing a proof-of-concept for a big data system. Which phase of the big * A. Engage
data adoption cycle is the company currently in? B. Execute
C. Explore
D. Educate
Which feature in a Big SQL federation is a library to access a particular type of data source? A. server
B. table
C. view
* D. wrapper
What is a feature of Apache ZooKeeper? A. generates shell programs for running components of Hadoop
B. monitors log files of cluster members
* C. maintains configuration information for a cluster
D. performance tunes a running cluster

Which open source component is a big data processing framework?

* A. Apache Spark
B. Apache Ambari
C. IBM BigSheets
D. IBM Big SQL
Which command is used to launch an interactive Python shell for Spark? A. python -spark
* B. pyspark
C. hadoop pyshell
D. spark-shell
What are the two types of Spark operations? (Choose two.) A. Sequences
(Please select ALL that apply) B. Vectors
C. DataFrames
* D. Transformations
* E. Actions
Which statement is true regarding Reduce tasks in MapReduce?
* A. They can run on any node.
B. They only run on nodes that didn't generate data during the Map step.
C. They run only on nodes that generated data during the Map step.
D. They only run on one node.
What command will load the BigR package in R? * A. library(bigr)
B. source("bigr")
C. dir(pattern="bigr")
D. bigr.connect
Which programming language is Apache Spark primarily written in? * A. Scala
B. Java
C. Python 2
D. C++
Which feature of Text Analytics should you use to process Japanese or Chinese language text? A. Annotation Query Language (AQL)
B. Standard tokenizer
* C. Multilingual tokenizer
D. Online Analytical Programming (OLAP)
Which kind of HBase row key maps to multiple SQL columns?
A. Primary
B. Dense
C. Unique
* D. Composite
What does the HCatalog component of Hive provide? A. collecting common data transformations into a library
B. maintaining an inventory of cluster nodes
* C. table and storage management layer for Hadoop
D. providing a REST gateway for jobs
Which action is performed prior to the Map step of a MapReduce v1 processing cycle? A. The job is sent sequentially to all nodes.
B. Output result sets are simplified to a single answer.
C. The data required is moved to the fastest nodes.
* D. The job is broken into individual task pieces and distributed.
Which integration API does Apache Ambari support? * A. REST
B. RMI
C. SOAP
D. RPC
How does an end-user interact with the IBM BigSheets tool? A. IBM-built desktop app
B. command line
C. mobile app
* D. web browser
Which software is at the core of the IBM BigInsights platform? * A. open source components
B. customer developed software
C. proprietary IBM libraries
D. cloud-based web services
What command is used to retrieve multiple rows out of an HBase table? A. pull
B. select
* C. scan
D. get
Which format is used to export extractor results? A. TXT
B. RTF
C. JSON
* D. CSV
How does Sqoop decide how to split data across mappers? * A. examining the primary key
B. moving the data to the closest network node
C. dividing the input bytes by available nodes
D. applying the split size to the data
How should you use the pre-built extractors for a new project? A. Assign them to a new query.
B. Right click on the extractor and select Edit Output.
* C. Drag and drop them onto canvas.
D. Convert them to AQL Statements.
Why is the SYSPROC.SYSINSTALLOBJECT procedure used with Big SQL? A. to create a SNAPSHOT column
B. to set the location of the EXPLAIN.DDL
C. to specify the SQL statement to be explained
* D. to create an EXPLAIN table
What does the programmatic implementation of a Map function do? * A. Reads the data file and performs a transformation.
B. Combines previous results into an aggregate.
C. Locates the data in the DFS.
D. Computes the final result of the entire job.
Which statement will create a table with parquet files? * A. CREATE HADOOP TABLE T ( i int, s VARCHAR(10)) STORED AS PARQUETFILE;.
B. CREATE HADOOP TABLE T ( i int, s VARCHAR(10)) STORED AS PARQUET;
C. CREATE HADOOP TABLE T ( i int, s VARCHAR(10)) SAVE AS PARQUETFILE;
D. CREATE HADOOP TABLE T ( i int, s VARCHAR(10)) SAVE AS PARQUET;
What is the JSqsh tool used for? A. web-based SQL editing
B. deploying the SQL JDBC driver
* C. command-line SQL queries
D. installing the IBM Data Server Manager (DSM)

What does the bucketing feature of Hive do? * A. sub-partitioning/grouping of data by hash within partitions
B. allows data to be stored in arrays
C. splits data into collections based on ranges
D. distributes the data dynamically for faster processing
What advantage does the Text Analytics Web UI give you? * A. It generates the AQL syntax for you.
B. It allows only single data types.
C. It allows only one type of file extension.
D. It teaches you how to write AQL syntax.
Which AQL candidate rule combines tuples from two views with the same schema? A. Blocks
B. Select
* C. Union
D. Sequence
Data collected within your organization has a short period of time when it is relevant. Which * A. Velocity
characteristic of a big data system does this represent? B. Validation
C. Variety
D. Volume
Assuming the same data is stored in multiple data formats, which format will provide faster * A. Parquet
query execution and require the least amount of IO operations to process? B. XML
C. flat file
D. JSON
Which feature of Text Analytics allows you to rollback your extractors when necessary? * A. Snapshots
B. Standard tokenizer
C. Scalar functions
D. Multilingual tokenizer
What defines a relation in an AQL extractor? * A. a view
B. a row
C. a schema
D. a column
Which command must be run after compiling a Java program so it can run on the Hadoop * A. jar cf name.jar *.class
cluster? B. hadoop classpath
C. jar tf name.jar
D. rm hadoop.class
What type of NoSQL datastore does HBase fall into? A. document
B. key-value
* C. column
D. graph
Which data inconsistency may appear while using ZooKeeper? A. excessively stale data views
* B. simultaneously inconsistent cross-client views
C. unreliable client updates across the cluster
D. out-of-order updates across clients
What is required to run an EXPLAIN statement in Big SQL? A. the explainable-sql-statement clause
B. the SYSPROC.SYSINSTALLOBJECT procedure
* C. proper authorization
D. a rule

Which command must be run first to become the HDFS user? * A. su - hdfs
B. hadoop fs
C. pwd
D. hdfs
Which Big SQL file format is human readable and supported by most tools, but is the least * A. Delimited
efficient file format? B. Parquet
C. Sequence
D. Avro
What is the default install location for the IBM Open Data Platform on Linux? A. /opt/ibm/iop
B. /var/iop
C. /usr/local/iop
* D. /usr/iop
You need to populate a Big SQL table to test an operation. Which INSERT statement is A. INSERT INTO ... SELECT FROM ...
recommended for testing, only because it does not support parallel reads or writes? * B. INSERT INTO ... VALUES (...)
C. INSERT INTO ... SELECT …
D. INSERT INTO ... SELECT ... WHERE …

Which command is used to launch an interactive Apache Spark shell? A. scala --spark
B. hadoop spark
C. spark
* D. spark-shell
Which data inconsistency may appear while using ZooKeeper? A. excessively stale data views
B. simultaneously inconsistent cross-client views
C. out-of-order updates across clients
D. unreliable client updates across the cluster
Which statement will create a table with parquet files? A. CREATE HADOOP TABLE T ( i int, s VARCHAR(10)) STORED AS PARQUET;
B. CREATE HADOOP TABLE T ( i int, s VARCHAR(10)) SAVE AS PARQUETFILE;
* C. CREATE HADOOP TABLE T ( i int, s VARCHAR(10)) STORED AS PARQUETFILE;.
D. CREATE HADOOP TABLE T ( i int, s VARCHAR(10)) SAVE AS PARQUET;
What are extractors transformed into when they are executed? A. Candidate generation statements
B. BigSheets function statements
* C. Annotated Query Language (AQL) statements
D. Online Analytical Programming (OLAP) statements
You need to set up the command-line interface JSqsh to connect to a bigsql database. What is A. Run the $JSQSH_HOME/bin/JSQSH script.
the recommended method to set up the connection? B. Run the JSqsh driver wizard.
C. Modify database parameters in the .jsqsh/connections.xml file.
* D. Run the JSqsh connection wizard.
How will the following column mapping command be encoded? cf_data:full_names mapped A. Hex
by (last_name, First_name) separator ',' B. Character
C. Binary
* D. String
Which underlying data representation and access method does Big SQL use? * A. Hive
B. TINYINT
C. MAP
D. SMALLINT
What does the MLlib component of Apache Spark support? * A. scalable machine learning
B. graph computation
C. SQL and HiveQL
D. stream processing

Which type of HBase column is mapped to multiple SQL columns? A. Composite

B. Exclusive
* C. Dense
D. Double
What is a key factor in determining how to implement file compression with HDFS? * A. compression algorithm supports splitting
B. the CPU speed of the cluster members (MHz)
C. the speed of network transfers between nodes
D. the amount of storage space needed for all files

What is used in a Big SQL file system to organize tables? A. JSqsh

B. DSM
* C. schemas
D. partitions
What command is used to start a Flume agent? A. flume-start
B. flume-src
* C. flume-ng
D. flume-agent
Which component is required for Flume to work? A. RDBMS
B. Syslog
* C. Data source
D. Interceptor

When creating a new table in Big SQL, what additional keyword is used in the CREATE TABLE A. dfs
statement to create the table in HDFS? * B. hadoop
C. replicated
D. cloud

What is a feature of an Avro file? A. directly readable by JavaScript

* B. versioning of the data
C. columns delimited by commas
D. formal schema language

What does the federation feature of Big SQL allow? A. tuning server hardware performance
B. importing data into HDFS
C. rewriting statements for better execution performance
* D. querying multiple data sources in one statement
A Hadoop file listing is performed and one of the output lines is: -rw-r--r-- 5 biadmin biadmin A. permissions
871233 2015-09-12 09:33 data.txt What does the 5 in the output represent? * B. replication factor
C. login id of the file owner
D. data size

What privilege is required to execute an EXPLAIN statement with INSERT privileges in Big SQL? A. SYSMON authority
B. SECADM authority
* C. SQLADM authority
D. SYSCTRL authority

What does a computer need to understand unstructured data? * A. context

B. usage
C. extractors
D. attribute types
What is the ApplicationMaster in YARN responsible for? (Choose two.) * A. monitoring node execution status
(Please select ALL that apply) * B. obtaining resources for computation
C. taking nodes offline for maintenance
D. allocating resources from all nodes

What is a limitation of Apache Spark? * A. It does not have universal tools.

B. It does not support streams.
C. It does not run Hadoop.
D. It does not in itself interact with SQL.
How is a sequence created in Canvas? A. Click on the New Literal button.
* B. Drag and drop one extractor onto another.
C. Right click on the extractor, and select Edit Output.
D. Select multiple extractors on the result pane.
Which type of key does HBase require in each row in an HBase table? A. Duplicate
B. Foreign
* C. Unique
D. Primary
What is a key factor in determining how to implement file compression with HDFS? * A. compression algorithm supports splitting
B. the speed of network transfers between nodes
C. the amount of storage space needed for all files
D. the CPU speed of the cluster members (MHz)
Which two components make up a Hadoop node? (Choose two.) * A. CPU
(Please select ALL that apply) B. network
C. memory
* D. disk
Which statement is used to set the correct compatible collation with Big SQL? * A. CREATE SERVER
B. SEQUENCE
C. PUSHDOWN
D. CREATE WRAPPER
How does Apache Ambari use the Ganglia component? A. to predict hardware failures
* B. to monitor cluster performance
C. to cluster job scheduling
D. to add new nodes to the cluster

How can you fix duplicate results generated by an extractor from the same text because the A. edit output with overlapping matches
text matches more than one dictionary entry? B. remove union statement
* C. remove with a consolidation rule
D. edit properties of the sequence
Which statement best describes Spark? A. An instance of a federated database.
* B. A computing engine for a large-scale data set.
C. A logical view on top of Hadoop data.
D. An open source database query tool.
Which two tasks can an Apache Ambari admin do that a regular Apache Ambari user cannot A. browse job information
do? (Choose two.) B. view service status
(Please select ALL that apply) * C. modify configurations
* D. run service checks
What is the default replication factor for HDFS on a production cluster? A. 5
B. 1
C. 10
* D. 3
In the ZooKeeper environment, what does atomicity guarantee? * A. Updates completely succeed or fail.
B. Updates are applied in the order created.
C. If an update succeeds, then it persists.
D. Every client sees the same view.
Which basic feature rule of AQL helps find an exact match to a single word or phrase? A. Dictionary
B. Part of Speech
* C. Literals
D. Splits

What is the default data type in Big R? * A. character

B. integer
C. complex
D. numeric

You have a very large Hadoop file system. You need to work on the data without migrating the A. MapReduce
data out or changing the data format. Which IBM tool should you use? * B. Big SQL
C. Data Server Manager
D. Pig
Which core component of the Hadoop framework is highly scalable and a common tool? A. Sqoop
B. Pig
* C. MapReduce
D. Hive
How can you reduce the memory usage of the ANALYZE command in Big SQL? A. Run everything in one batch.
B. Turn on distribution statistics.
* C. Run the command separately on different batches of columns.
D. Include all the columns in the batch.
What should you do in Text Analytics to fix an extractor that produces unwanted results? A. Re-create the extractors.
B. Remove results with a consolidation rule.
* C. Create a new filter.
D. Edit the properties of the sequence.
QUESTIONS REPONSES CORRECTION
A The list of deployments
B A list of your saved bookmarks
C The email address of the collaborator
You need to add a collaborator to your project. What do you need? D Your project ID c
A URL
B Scala
C File
Before you create a Jupyter notebook in Watson Studio, which two items are necessary? D Spark Instance
(Please select the TWO that apply) E Project de
 A Database
B Wrapper
C Object Storage
Where does the unstructured data of a project reside in Watson Studio? D Tables c
A Watson Studio Desktop
B Watson Studio Cloud
C Watson Studio Business
Which Watson Studio offering used to be available through something known as IBM Bluemix? D Watson Studio Local b
A Data Assets
B Projects
C Collaborators
What is the architecture of Watson Studio centered on? D Analytic Assets b
A INSERT
B GRANT
C REVOKE
Which two commands would you use to give or remove certain privileges to/from a user? D LOAD
(Please select the TWO that apply) E SELECT bc
A4
B2
C1
How many Big SQL management node do you need at minimum? D3 c
A /apps/hive/warehouse/data
B /apps/hive/warehouse/bigsql
C /apps/hive/warehouse/
What is the default directory in HDFS where tables are stored? D /apps/hive/warehouse/schema c
A ./java mybigdata
B ./jsqsh mybigdata
C ./java tables
Using the Java SQL Shell, which command will connect to a database called mybigdata? D ./jsqsh go mybigdata b
A 777
B 755
C 700
Which directory permissions need to be set to allow all users to create their own schema? D 666 a
A Files
B Schemas
C Hives
What are Big SQL database tables organized into? D Directories b
A hdfs dfs -chmod 770 /hive/warehouse
B hdfs dfs -chmod 755 /hive/warehouse
C hdfs dfs -chmod 700 /hive/warehouse
You have a distributed file system (DFS) and need to set permissions on the the /hive/warehouseDdirectory
hdfs dfs to-chmod
allow 666
access
/hive/warehouse
to ONLY the bigsql user. Which command would you run? c
A It grants or revokes certain user privileges.
B It grants or revokes certain directory privileges.
C It limits the rows or columns returned based on certain criteria.
Which definition best describes RCAC? D It limits access by using views and stored procedures. c
A A data type of a column describing its value.
B The defined format and rules around a delimited file.
C A container for any record format.
Which statement best describes a Big SQL database table? D A directory with zero or more data files. d
A Big SQL can exploit advanced features
B Data interchange outside Hadoop
C Supported by multiple I/O engines
What is an advantage of the ORC file format? D Efficient compression d
 A GRANT
B umask
C Kerberos
You need to determine the permission setting for a new schema directory. Which tool would youDuse?HDFS b
A Scheduler
B DSM
C Jupyter
Which tool would you use to create a connection to your Big SQL database? D Ambari b
A bigsql.alltables.io.doAs
B DB2COMPOPT
C DB2_ATS_ENABLE
You need to enable impersonation. Which two properties in the bigsql-conf.xml file need to be marked true?
D $BIGSQL_HOME/conf
(Please select the TWO that apply) E bigsql.impersonation.create.table.grant.public ae
A CREATE AS parquet
B STORED AS parquetfile
C STORED AS parquet
You are creating a new table and need to format it with parquet. Which partial SQL statement would
D CREATE
createAS
the
parquetfile
table in parquet format? b
A TRANSLATE FUNCTION
B CREATE FUNCTION
C ALTER MODULE ADD FUNCTION
Which command creates a user-defined schema function? D ALTER MODULE PUBLISH FUNCTION b
A YARN
B Spark
C Pig
Which Apache Hadoop application provides an SQL-like interface to allow abstraction of data on semi-structured
D Hive data in a Hadoop datastore? d
A A messaging system for real-time data pipelines.
B A wizard for installing Hadoop services on host servers.
C Moves information to/from structured databases.
Which description characterizes a function provided by Apache Ambari? D Moves large amounts of streaming event data. b
A Writes to a leader server will always succeed.
B All servers keep a copy of the shared data in memory.
C There can be more than one leader server at a time.
Which statement accurately describes how ZooKeeper works? D Clients connect to multiple servers at the same time. b
A MemcacheD
B CouchDB
C Riak
Which NoSQL datastore type began as an implementation of Google's BigTable that can store anyDtypeHbase of data and scale to many petabytes? d
A It is a powerful platform for managing large volumes of structured data.
B It is designed specifically for IBM Big Data customers.
C It is a Hadoop distribution based on a centralized architecture with YARN at its core.
Which statement is true about Hortonworks Data Platform (HDP)? D It is engineered and developed by IBM's BigInsights team. c
A Pig
B Hive
C Python
What is the name of the Hadoop-related Apache project that utilizes an in-memory architecture to
D Spark
run applications faster than MapReduce? d
A It runs on Hadoop clusters with RAM drives configured on each DataNode.
B It supports HDFS, MS-SQL, and Oracle.
C It is much faster than MapReduce for complex applications on disk.
Which statement about Apache Spark is true? D It features APIs for C++ and .NET. c
A It determines the size and distribution of data split in the Map phase.
B It aggregates all input data before it goes through the Map phase.
C It reduces the amount of data that is sent to the Reducer task nodes.
Which statement is true about the Combiner phase of the MapReduce architecture? D It is performed after the Reducer phase to produce the final output. c
 A MapReduce v1 APIs cannot be used with YARN.
B MapReduce v1 APIs provide a flexible execution environment to run MapReduce.
C MapReduce v1 APIs are implemented by applications which are largely independent of the execution environment.
Which statement is true about MapReduce v1 APIs? D MapReduce v1 APIs define how MapReduce jobs are executed. c
A Hive
B Cloudbreak
C Big SQL
D MapReduce
Hadoop 2 consists of which three open-source sub-projects maintained by the Apache Software Foundation?
E HDFS
(Please select the THREE that apply) F YARN def
A Authorization Provider
B Ambari Alert Framework
C Postgres RDBMS
Which component of the Apache Ambari architecture integrates with an organization's LDAP or Active
D RESTDirectory
API service? a
A Authenticating and auditing user access.
B Loading bulk data into an Hadoop cluster.
What are two services provided by ZooKeeper? C Maintaining configuration information.
(Please select the TWO that apply) D Providing distributed synchronization. cd
A Sesame
B Neo4j
C MongoDB
What is an example of a Key-value type of NoSQL datastore? D REDIS d
A MLlib
B RDD
C Mesos
Which Spark Core function provides the main element of Spark API? D YARN b
A org.apache.mr
B org.apache.hadoop.mr
C org.apache.hadoop.mapred
Which is the java class prefix for the MapReduce v1 APIs? D org.apache.mapreduce c
A SSD
B JBOD
C RAID
Which hardware feature on an Hadoop datanode is recommended for cost efficient performance? D LVM b
A ApplicationMaster
B JobMaster
C TaskManager
Under the YARN/MRv2 framework, the JobTracker functions are split into which two daemons? D ScheduleManager
(Please select the TWO that apply) E ResourceManager ae
A The number of rows to commit per transaction.
B The number of rows to send to each mapper.
C The table name to export from the database.
What does the split-by parameter tell Sqoop? D The column to use as the primary key. d
A Ambari
B Google File System
C HBase
Hadoop uses which two Google technologies as its foundation? D YARN
(Please select the TWO that apply) E MapReduce be
A Requires extremely rapid processing.
B Data is processed in batch.
Which two are attributes of streaming data? C Simple, numeric data.
(Please select the TWO that apply) D Sent in high volume. ad
Which statement accurately describes how ZooKeeper works? All servers keep a copy of the shared data in memory.
 SequenceFiles
A API and perimeter security.
B Management of Kerberos in the cluster.
What two security functions does Apache Knox provide? C Proxying services.
(Please select the TWO that apply) D Database field access auditing. ac
A REDIS
B HBase
C Cassandra
What is an example of a NoSQL datastore of the "Document Store" type? D MongoDB d
A Hive
B Sqoop
C Pig
Which Apache Hadoop application provides a high-level programming language for data transformation
D Zookeeper
on unstructured data? c
A Big Data
B Big Match
C Big Replicate
D Big SQL
What are three IBM value-add components to the Hortonworks Data Platform (HDP)? E Big YARN
(Please select the THREE that apply) F Big Index bcd
A Accumulo
B HBase
C Oozie
Which Hadoop ecosystem tool can import data into a Hadoop cluster from a DB2, MySQL, or otherD Sqoop
databases? d
A Data Protection
B Speed
C Resiliency
Which three are a part of the Five Pillars of Security? D Audit
(Please select the THREE that apply) E Administration ade
A MLlib
B Mesos
C Spark SQL
Which component of the Spark Unified Stack allows developers to intermix structured database queries
D Java with Spark's programming language? c
A Hadoop YARN
B Apache Mesos
C Nomad
Apache Spark can run on which two of the following cluster managers? D Linux Cluster Manager
(Please select the TWO that apply) E oneSIS ab
A Suitable for transaction processing.
B Libraries that support SQL queries.
C APIs for Scala, Python, C++, and .NET.
Which feature makes Apache Spark much easier to use than MapReduce? D Applications run in-memory. b
A Run Sqoop using the vi editor.
B Use the --import-command line argument.
What are two ways the command-line parameters for a Sqoop invocation can be simplified? C Include the --options-file command line argument.
(Please select the TWO that apply) D Place the commands in a file. cd
A NodeChildrenChanged
B NodeDeleted
Which two are valid watches for ZNodes in ZooKeeper? C NodeExpired
(Please select the TWO that apply) D NodeRefreshed ab
A ResourceManager
B JobMaster
C ScheduleManager
Under the YARN/MRv2 framework, which daemon arbitrates the execution of tasks among all theDapplications
ApplicationMaster
in the system? a
 A NiFi
B Hortonworks Data Flow
C Druid
What is the preferred replacement for Flume? D Storm b
A Use the -mapper 1 parameter.
B Use the --limit mapper=1 parameter.
C Use the -m 1 parameter.
How can a Sqoop invocation be constrained to only run one mapper? D Use the --single parameter. c
A Reduce
B Map
C Combiner
Under the MapReduce v1 programming model, which optional phase is executed simultaneouslyDwith Splitthe Shuffle phase? c
A Acquisition
B Manipulation
C Exploration
What is the first step in a data science pipeline? D Analytics a
A Holding the output of a computation.
B Configuring data connections.
C Documenting the computational process.
What is a markdown cell used for in a data science notebook? D Writing code to transform data. c
A Common desktop app.
B Database interface.
C Linux SSH session.
What does the user interface for Jupyter look like to a user? D App in web browser. d
A To display a simple bar chart of data on the screen.
B To collect video for use in streaming data applications.
C To perform certain data transformation quickly.
Why might a data scientist need a particular kind of GPU (graphics processing unit)? D To input commands to a data science notebook. c
A %dirmagic
B %lsmagic
C %list-magic
What command is used to list the "magic" commands in Jupyter? D %list-all-magic b
A The list of deployments
B A list of your saved bookmarks
You need to add a collaborator to your project. What do you need? C The email address of the collaborator
Your answer D Your project ID c
A record locking
B batch processing
C machine learning
D transaction processing
Apache Spark provides a single, unifying platform for which three of the following types of operations?
E graph operations
(Please select the THREE that apply) F ACID transactions bce
A Scala
B Python
C Java
D .NET
Which three programming languages are directly supported by Apache Spark? E C#
(Please select the THREE that apply) F C++ a / c/b
 A Map -> Split -> Reduce -> Combine

B Map -> Combine -> Shuffle -> Reduce

C Map -> Combine -> Reduce -> Shuffle

Under the MapReduce v1 programming model, which shows the proper order of the full set of MapReduce phases?
D Split -> Map -> Combine -> Reduce b
A ResourceManager
B JobMaster
C ApplicationMaster
Under the YARN/MRv2 framework, which daemon is tasked with negotiating with the NodeManager(s) D TaskManager
to execute and monitor tasks? c
A Collector
B Source
C Stream
What is the final agent in a Flume chain named? D Agent a
A Availability
B Authorization
What are two security features Apache Ranger provides? C Authentication
(Please select the TWO that apply) D Auditing bd
A Worker nodes store results on their own local file systems.
B Data is aggregated by worker nodes.
C Worker nodes process pieces in parallel.
Under the MapReduce v1 programming model, what happens in a "Reduce" step? D Input is split into pieces. b
A MapReduce
B Spark
C HBase
Which Apache Hadoop component can potentially replace an RDBMS as a large Hadoop datastoreD and Ambari
is particularly good for "sparse data"? c
A RCFile
B SequenceFiles
C Flat
Which data encoding format supports exact storage of all data in binary representations such as VARBINARY
D Parquet columns? b
A One time export and import of a database.
B An application evaluating sensor data in real-time.
C A system that stores many records in a database.
Which statement describes an example of an application using streaming data? D A web application that supports 10,000 users. b
A Scalability
B Resource utilization
C TaskTrackers can be a bottleneck to MapReduce jobs
What are two primary limitations of MapReduce v1? D Number of TaskTrackers limited to 1,000
(Please select the TWO that apply E Workloads limited to MapReduce ab
A RAM
B network
C CPU
Which component of an Hadoop system is the primary cause of poor performance? D disk latency d
A Partial failure of the nodes during execution.
B Finding a particular node within the cluster.
What are two common issues in distributed systems? C Reduced performance when compared to a single server.
(Please select the TWO that apply) D Distributed systems are harder to scale up. ab
A Ambari Metrics System
B Ambari Alert Framework
C Ambari Server
Which component of the Apache Ambari architecture provides statistical data to the dashboard about
D Ambari
the performance
Wizard of a Hadoop cluster? a
 A Impersonation
B Grant/Revoke privileges
C Fluid query
Which Big SQL feature allows users to join a Hadoop data set to data in external databases? D Integration c
A Data source
B User mapping
C Nickname
When connecting to an external database in a federation, you need to use the correct database driver
D Wrapper
and protocol. What is this federation component called in Big SQL? d
A Parsing and loading data into a notebook.
B Autoconfiguring data connections using a registry.
C Extending the core language with shortcuts.
What is a "magic" command used for in Jupyter? D Running common statistical analyses. c
A RAID-0
B Online Transactional Processing
C Parallel Processing
Which computing technology provides Hadoop's high performance D Online Analytical Processing c
A ScheduleManager
B ResourceManager
C ApplicationMaster
Under the YARN/MRv2 framework, the Scheduler and ApplicationsManager are components of which D TaskManager
daemon? b
A large number of small data files
B solid state disks
C immediate failover of failed disks
D high-speed networking between nodes
which two factors in a Hadoop cluster increase performance most significantly? E data redundancy on management nodes
(Please select the TWO that apply) F parallel reading of large data files ok d/F
A MapReduce
B YARN
C HDFS
Which component of the Hortonworks Data Platform (HDP) is the architectural center of HadoopDand Hbase
provides resource management and a central platform for Hadoop applications? b
A Ambari Alert Framework
B Ambari Wizard
C Ambari Metrics System
If a Hadoop node goes down, which Ambari component will notify the Administrator? D REST API a
A Code
B Kernel
C Output
Which type of cell can be used to document and comment on a process in a Jupyter notebook? D Markdown d
A Notebooks can be used by multiple people at the same time.
B Users must authenticate before using a notebook.
C Notebooks can be connected to big data engines such as Spark.
Which is an advantage that Zeppelin holds over Jupyter? D Zeppelin is able to use the R language. a
A Jaql
B Fault tolerance through HDFS replication
C Adaptive MapReduce
Which capability does IBM BigInsights add to enrich Hadoop? D Parallel computing on commodity servers c
A value
B volume
C verifiability
What is one of the four characteristics of Big Data? D volatility b
A Hadoop Common
B Hadoop HBase
C MapReduce
Which Hadoop-related project provides common utilities and libraries that support other HadoopDsubBigTable
projects? a
 A Federated Discovery and Navigation
B Text Analysis
C Stream Computing
Which type of Big Data analysis involves the processing of extremely large volumes of constantly moving
D MapReduce
data that is impractical to store? c
A 64-bit architecture
B disk latency
C MIPS
Which primary computing bottleneck of modern computers is addressed by Hadoop? D limited disk capacity b
A stream computing
B data warehousing
C analytics
Which Big Data function improves the decision-making capabilities of organizations by enabling the D distributed
organizations
fileto
system
interpret and evaluate structured and unstructured data in search of valuable
c business information?
A HBase
B Apache
C Jaql
What is one of the two technologies that Hadoop uses as its foundation? D MapReduce d
A a high throughput, shared file system
B high availability of the NameNode
C data access performed by an RDBMS
What key feature does HDFS 2.0 provide that HDFS does not? D random access to data in the cluster b
A LOAD
B JOIN
C TOP
What are two of the core operators that can be used in a Jaql query? (Select two.) D SELECT bc
A SQL-like
B compiled language
C object oriented
Which type of language is Pig? D data flow d
A hdfs.conf
B hadoop-configuration.xml
C hadoop.conf
If you need to change the replication factor or increase the default storage block size, which file do
D you
hdfs-site.xml
need to modify? d
A The file(s) must be stored on the local file system where the map reduce job was developed.
B The file(s) must be stored in HDFS or GPFS.
C The file(s) must be stored on the JobTracker.
To run a MapReduce job on the BigInsights cluster, which statement about the input file(s) must be D No
true?
matter where the input files are before, they will be automatically copied to where the
b job runs.
A operating system independence
B posix compliance
C no single point of failure
What is a characteristic of IBM GPFS that distinguishes it from other distributed file systems? D blocks that are stored on different nodes b
A Pig is used for creating MapReduce programs.
B Pig has a shelf interface for executing commands.
C Pig is not designed for random reads/writes or low-latency queries.
Which statement represents a difference between Pig and Hive? D Pig uses Load, Transform, and Store. d
A hdfs -dir mydata
B hadoop fs -mkdir mydata
C hadoop fs -dir mydata
Which command helps you create a directory called mydata on HDFS? D mkdir mydata b
A Reduce
B Shuffle
C Combine
In which step of a MapReduce job is the output stored on the local disk? D Map d
 A Worker nodes process individual data segments in parallel.
B Worker nodes store results in the local file system.
C Input data is split into smaller pieces.
Under the MapReduce programming model, which task is performed by the Reduce step? D Data is aggregated by worker nodes. d
A Reducer
B JobScheduler
C TaskTracker
Which element of the MapReduce architecture runs map and reduce jobs? D JobTracker c
A spread data across a cluster of computers
B provide structure to unstructured or semi-structured data
C increase storage capacity through advanced compression algorithms
What is one of the two driving principles of MapReduce? D provide a platform for highly efficient transaction processing a
A Cluster
B Distributed
C Remote
D Debugging
When running a MapReduce job from Eclipse, which BigInsights execution models are available? (Select
E Localtwo.) ae
A The number of reducers is always equal to the number of mappers.
B The number of mappers and reducers can be configured by modifying the mapred-site.xml file.
C The number of mappers and reducers is decided by the NameNode.
Which statement is true regarding the number of mappers and reducers configured in a cluster? D The number of mappers must be equal to the number of nodes in a cluster. b
A hadoop size
B hdfs -du
C hdfs fs size
Which command displays the sizes of files and directories contained in the given directory, or the Dlength
hadoop
of afsfile,
-duin case it is just a file? d
A three
B two
C one
Following the most common HDFS replica placement policy, when the replication factor is three, how D none
many replicas will be located on the local rack? c
A copies Job Resources to the shared file system
B coordinates the job execution
C executes the map and reduce functions
In the MapReduce processing model, what is the main function performed by the JobTracker? D assigns tasks to each cluster node b
A Both are data flow languages.
B Both require schema.
C Both use Jaql query language.
How are Pig and Jaql query languages similar? D Both are developed primarily by IBM. a
A to manage storage attached to nodes
B to coordinate MapReduce jobs
C to regulate client access to files
Under the HDFS architecture, what is one purpose of the NameNode? D to periodically report status to DataNode c
A hadoop fs list
B hdfs root
C hadoop fs -Is /
Which command should be used to list the contents of the root directory in HDFS? D hdfs list / c
A runs map and reduce tasks
B keeps the work physically close to the data
C reports status of DataNodes
What is one function of the JobTracker in MapReduce? D manages storage b
A built-in UDFs and indexing
B platform-specific SQL libraries
C an RDBMS such as DB2 or MySQL
In addition to the high-level language Pig Latin, what is a primary component of the Apache Pig platform?
D runtime environment d
 A Data is accessed through MapReduce.
B Data is designed for random access read/write.
C Data can be processed over long distances without a decrease in performance.
Which statement is true about Hadoop Distributed File System (HDFS)? D Data can be created, updated and deleted. a
A managing customer information in a CRM database
B sentiment analytics from social media blogs
C product cost analysis from accounting systems
Which is a use-case for Text Analytics? D health insurance cost/benefit analysis from payroll data b
A BigSheets client
B Microsoft Excel
C Eclipse
Which tool is used to access BigSheets? D Web Browser d
A Hive metastore
B RDBMS
C MapReduce
Which technology does Big SQL utilize for access to shared catalogs? D HCatalog a
A display view <view_name>
B return view <view_name>
C output view <view_name>
Which statement will make an AQL view have content displayed? D export view <view_name> c
A Text Analytics
B Stream Computing
C Data Warehousing
You work for a hosting company that has data centers spread across North America. You are trying
D Temporal
to resolveAnalysis
a critical performance problem in which a large number of web servers are performing
a far below expectations. You know that the infor
A Thrift client
B Hive shell
C Hive SQL client
Which utility provides a command-line interface for Hive? D Hive Eclipse plugin b
A It is a data flow language for structured data based on Ansi-SQL.
B It is a distributed file system that replicates data across a cluster.
C It is an open source implementation of Google's BigTable.
What is an accurate description of HBase? D It is a database schema for unstructured Big Data. c
A BigSQL
B BigSheets
C Avro
Which Hadoop-related technology provides a user-friendly interface, which enables business users
D HBase
to easily analyze Big Data? b
A Text Analytics is the most common way to derive value from Big Data.
B MapReduce is unable to process unstructured text.
C Data warehouses contain potentially valuable information.
What drives the demand for Text Analytics? D Most of the world's data is in unstructured or semi-structured text. d
A An external table refers an existing location outside the warehouse directory.
B An external table refers to a table that cannot be dropped.
C An external table refers to the data from a remote database.
In Hive, what is the difference between an external table and a Hive managed table? D An external table refers to the data stored on the local file system. a
A It provides all the capabilities of an RDBMS plus the ability to manage Big Data.
B It is a database technology that does not use the traditional relational model.
C It is based on the highly scalable Google Compute Engine.
Which statement about NoSQL is true? D It is an IBM project designed to enable DB2 to manage Big Data. b
A "Copy" to create a new sheet with the other workbook data in the current workbook
B "Group" to bring together the two workbooks
C "Load" to create a new sheet with the other workbook data in the current workbook
If you need to JOIN data from two workbooks, which operation should be performed beforehand? D "Add" to add the other workbook data to the current workbook c
 A to get detailed information about the table
B to view data in an Hbase table
C to report any inconsistencies in the database
What is the "scan" command used for in HBase? D to list all tables in Hbase c
A Eclipse with BigInsights tools for Eclipse plugin
B BigInsights Console with AQL plugin
C AQLBuilder
Which tool is used for developing a BigInsights Text Analytics extractor? D AQL command line a
A Pre-create regions by specifying splits in create table command and use the insert command to load data.
B Pre-create regions by specifying splits in create table command and bulk loading the data.
C Pre-create the column families when creating the table and use the put command to load the data.
What is the most efficient way to load 700MB of data when you create a new HBase table? D Pre-create the column families when creating the table and bulk loading the data. b
The following sequence of commands is executed:
create 'table_1','column_family1','column_family2'
put 'table_1','row1','column_family1:c11','r1v11'
put 'table_1','row2','column_family1:c12','r1v12'
put 'table_1','row2','column_family2:c21','r1v21'
put 'table_1','row3','column_family1:d11','r1v11' A4
put 'table_1','row2','column_family1:d12','r1v12' B3
put 'table_1','row2','column_family2:d21','r1v21' C6
In HBase, which value will the "count 'table_1'" command return? D2 b
A TRANSFORM
B SELECT
C GET
Which Hive command is used to query a table? D EXPAND b
A because SQL enhances query performance
B because the MapReduce Java API is sometimes difficult to use
C because data stored in a Hadoop cluster lends itself to structured SQL queries
Why develop SQL-based query languages that can access Hadoop data sets? D because the data stored in Hadoop is always structured b
A It allows Hadoop to apply the schema-on-ingest model to unstructured Big Data.
B It allows an RDBMS to maintain referential integrity on a Hadoop data set.
C It allows customers to leverage high-end server platforms to manage Big Data.
Which key benefit does NoSQL provide? D It can cost-effectively manage data sets too large for traditional RDBMS. d
A Hadoop data is highly structured.
B Data is in many formats.
C Data is located on a distributed file system.
What makes SQL access to Hadoop data difficult? D Hadoop requires pre-defined schema. b
A list tables
B describe tables
C show all
Which command can be used in Hive to list the tables available in a database/schema? D show tables d
A to count the number of columns of a table
B to count the number of column families of a table
C to count the number of rows in a table
In HBase, what is the "count" command used for? D to count the number of regions of a table c
A HBase
B Pig
C Jaql
Which Hadoop-related technology supports analysis of large datasets stored in HDFS using an SQL-like
D Hivequery language? d
A They need to be marked as "Shared."
B They need to be copied under the user home directory.
C They need to be deployed with proper privileges.
How can the applications published to BigInsights Web Console be made available for users to execute?
D They need to be linked with the master application. c
 A Eclipse
B Oozie
C Jaql
Which component of Apache Hadoop is used for scheduling and running workflow jobs? D Task Launcher b
A validater
B replicater
C crawler
What is one of the main components of Watson Explorer (InfoSphere Data Explorer)? D compressor c
A analyze and react to data in motion before it is stored
B find and analyze historical stream data stored on disk
C analyze and summarize product sentiments posted to social media
IBM InfoSphere Streams is designed to accomplish which Big Data function? D execute ad-hoc queries against a Hadoop-based data warehouse c
A InfoSphere Information Server
B InfoSphere Streams
C InfoSphere BigInsights
Which IBM Big Data solution provides low-latency analytics for processing data-in-motion? D PureData for Analytics b
A Avro
B HBase
C Eclipse
Which IBM tool enables BigInsights users to develop, test and publish BigInsights applications? D BigInsights Applications Catalog c
A enabling customers to efficiently index and access large volumes of data
B gaining new insight through the capabilities of the world's interconnected intelligence
C providing solutions to help customers manage and grow large database systems
Which description identifies the real value of Big Data and Analytics? D using modern technology to efficiently store the massive amounts of data generated by social
b networks
Sqoop
Which Hadoop ecosystem tool can import data into a Hadoop cluster from a DB2, MySQL, or other databases?

HBase
Which NoSQL datastore type began as an implementation of Google's BigTable that can store any type of data and scale to many petabytes?
A Parallel Processing
B Online Analytical Processing
C Online Transactional Processing
Which computing technology provides Hadoop's high performance? D RAID-0 a
 Test Blanc

Question 1
You need to determine the permission setting for a
new schema directory. Which tool would you use?
Your answer

A. Kerberos

B. HDFS

C. umask

D. GRANT

Question 2
Using the Java SQL Shell, which command will
connect to a database called mybigdata?
Your answer

A. ./java tables

B. ./jsqsh mybigdata

C. ./jsqsh go mybigdata

D. ./java mybigdata
 Test Blanc

Question 3
Which tool would you use to create a connection to
your Big SQL database?
Your answer

A. Ambari

B. Scheduler

C. DSM

D. Jupyter

Which definition best describes RCAC?

Your answer

A. It limits the rows or columns returned based on certain criteria.

B. It grants or revokes certain directory privileges.

C. It grants or revokes certain user privileges.

D. It limits access by using views and stored procedures.

 Test Blanc
Question 5
What is an advantage of the ORC file format?
Your answer

A. Data interchange outside Hadoop

B. Supported by multiple I/O engines

C. Big SQL can exploit advanced features

D. Efficient compression

Question 6
You need to enable impersonation. Which two
properties in the bigsql-conf.xml file need to be
marked true?
Your answer

A. bigsql.alltables.io.doAs

B. bigsql.impersonation.create.table.grant.public

C. DB2_ATS_ENABLE

D. DB2COMPOPT

E. $BIGSQL_HOME/conf
 Test Blanc
Question 7
What are Big SQL database tables organized into?
Your answer

A. Directories

B. Files

C. Hives

D. Schemas

Question 8
Which two commands would you use to give or
remove certain privileges to/from a user?
Your answer

A. GRANT

B. REVOKE

C. INSERT

D. SELECT

E. LOAD
 Test Blanc

Question 9
You have a distributed file system (DFS) and need to
set permissions on the the /hive/warehouse
directory to allow access to ONLY the bigsql user.
Which command would you run?
Your answer

A. hdfs dfs -chmod 770 /hive/warehouse

B. hdfs dfs -chmod 700 /hive/warehouse

C. hdfs dfs -chmod 755 /hive/warehouse

D. hdfs dfs -chmod 666 /hive/warehouse

Which statement best describes a Big SQL database

table?
Your answer

A. A container for any record format.

B. The defined format and rules around a delimited file.

C. A directory with zero or more data files.

D. A data type of a column describing its value.

 Test Blanc

Question 11
Which Big SQL feature allows users to join a
Hadoop data set to data in external databases?
Your answer

A. Integration

B. Fluid query

C. Impersonation

D. Grant/Revoke privileges

Question 12
Which directory permissions need to be set to allow
all users to create their own schema?
Your answer

A. 700

B. 755

C. 666

D. 777
 Test Blanc

Question 13
What is the default directory in HDFS where tables
are stored?
Your answer

A. /apps/hive/warehouse/

B. /apps/hive/warehouse/data

C. /apps/hive/warehouse/bigsql

D. /apps/hive/warehouse/schema

Question 14
How many Big SQL management node do you need
at minimum?
Your answer

A. 1

B. 2

C. 4

D. 3
 Test Blanc
Question 15
When connecting to an external database in a
federation, you need to use the correct database
driver and protocol. What is this federation
component called in Big SQL?
Your answer

A. Wrapper

B. User mapping

C. Data source

D. Nickname

Question 16
Which Apache Hadoop application provides an SQL-
like interface to allow abstraction of data on semi-
structured data in a Hadoop datastore?
Your answer

A. Pig

B. Spark

C. Hive

D. YARN

Question 17
Apache Spark provides a single, unifying platform
for which three of the following types of operations?
 Test Blanc
Your answer

A. record locking

B. machine learning

C. batch processing

D. ACID transactions

E. graph operations

F. transaction processing

Question 18
Under the YARN/MRv2 framework, which daemon is
tasked with negotiating with the NodeManager(s) to
execute and monitor tasks?
Your answer

A. JobMaster

B. ApplicationMaster

C. ResourceManager

D. TaskManager

Question 19
 Test Blanc
Which component of the Hortonworks Data Platform
(HDP) is the architectural center of Hadoop and
provides resource management and a central
platform for Hadoop applications?
Your answer

A. MapReduce

B. HBase

C. YARN

D. HDFS

Question 20
What is the final agent in a Flume chain named?
Your answer

A. Agent

B. Stream

C. Collector

D. Source

Question 21
Which Apache Hadoop application provides a high-
level programming language for data transformation
on unstructured data?
Your answer

A. Sqoop
 Test Blanc
Your answer

B. Zookeeper

C. Hive

D. Pig

Question 22
Which component of the Spark Unified Stack allows
developers to intermix structured database queries
with Spark's programming language?
Your answer

A. Java

B. MLlib

C. Mesos

D. Spark SQL
 Test Blanc

Question 23
Under the MapReduce v1 programming model, what
happens in a "Reduce" step?
Your answer

A. Worker nodes process pieces in parallel.

B. Worker nodes store results on their own local file systems.

C. Input is split into pieces.

D. Data is aggregated by worker nodes.

Question 24
Which statement is true about Hortonworks Data
Platform (HDP)?
Your answer

A. It is designed specifically for IBM Big Data customers.

B. It is a powerful platform for managing large volumes of structured data.

C. It is a Hadoop distribution based on a centralized architecture with YARN at its core.

D. It is engineered and developed by IBM's BigInsights team.

Apache Spark can run on which two of the following

cluster managers?
 Test Blanc
Your answer

A. Nomad

B. Hadoop YARN

C. oneSIS

D. Apache Mesos

E. Linux Cluster Manager

Question 26
Which statement is true about the Combiner phase
of the MapReduce architecture?
Your answer

A. It determines the size and distribution of data split in the Map phase.

B. It reduces the amount of data that is sent to the Reducer task nodes.

C. It is performed after the Reducer phase to produce the final output.

D. It aggregates all input data before it goes through the Map phase.

Question 27
Which three are a part of the Five Pillars of
Security?
Your answer

A. Administration

B. Audit
 Test Blanc
Your answer

C. Speed

D. Data Protection

E. Resiliency

Question 28
Which feature makes Apache Spark much easier to
use than MapReduce?
Your answer

A. Libraries that support SQL queries.

B. Applications run in-memory.

C. Suitable for transaction processing.

D. APIs for Scala, Python, C++, and .NET.

Under the YARN/MRv2 framework, the JobTracker

functions are split into which two daemons?
Your answer

A. ResourceManager

B. TaskManager

C. JobMaster

D. ApplicationMaster
 Test Blanc
Your answer

E. ScheduleManager

Under the YARN/MRv2 framework, the JobTracker

functions are split into which two daemons?
Your answer

A. ResourceManager

B. TaskManager

C. JobMaster

D. ApplicationMaster

E. ScheduleManager

Question 30
Which computing technology provides Hadoop's
high performance?
Your answer

A. RAID-0

B. Parallel Processing

C. Online Analytical Processing

D. Online Transactional Processing

Question 31
What are two security features Apache Ranger
provides?
 Test Blanc
Your answer

A. Availability

B. Authorization

C. Authentication

D. Auditing

Question 32
Which two are attributes of streaming data?
Your answer

A. Simple, numeric data.

B. Sent in high volume.

C. Data is processed in batch.

D. Requires extremely rapid processing.

Question 33
Which statement describes an example of an
application using streaming data?
Your answer

A. An application evaluating sensor data in real-time.

B. A web application that supports 10,000 users.

 Test Blanc
Your answer

C. One time export and import of a database.

D. A system that stores many records in a database.

Question 34
Which statement is true about MapReduce v1 APIs?
Your answer

A. MapReduce v1 APIs are implemented by applications which are largely independent of th

B. MapReduce v1 APIs define how MapReduce jobs are executed.

C. MapReduce v1 APIs provide a flexible execution environment to run MapReduce.

D. MapReduce v1 APIs cannot be used with YARN.

Question 35
What two security functions does Apache Knox
provide?
Your answer

A. API and perimeter security.

B. Database field access auditing.

C. Proxying services.

D. Management of Kerberos in the cluster.

Question 36
 Test Blanc
Hadoop 2 consists of which three open-source sub-
projects maintained by the Apache Software
Foundation?
Your answer

A. HDFS

B. Hive

C. Big SQL

D. Cloudbreak

E. YARN

F. MapReduce

Question 37
Which Hadoop ecosystem tool can import data into
a Hadoop cluster from a DB2, MySQL, or other
databases?
Your answer

A. Sqoop

B. HBase

C. Accumulo

D. Oozie

Question 38
Which statement accurately describes how
ZooKeeper works?
 Test Blanc
Your answer

A. There can be more than one leader server at a time.

B. Writes to a leader server will always succeed.

C. All servers keep a copy of the shared data in memory.

D. Clients connect to multiple servers at the same time.

Question 39
What does the split-by parameter tell Sqoop?
Your answer

A. The number of rows to send to each mapper.

B. The column to use as the primary key.

C. The number of rows to commit per transaction.

D. The table name to export from the database.

Question 40
Which two are valid watches for ZNodes in
ZooKeeper?
Your answer

A. NodeChildrenChanged

B. NodeRefreshed

C. NodeDeleted
 Test Blanc
Your answer

D. NodeExpired

Question 41
Which component of the Apache Ambari
architecture integrates with an organization's LDAP
or Active Directory service?
Your answer

A. REST API

B. Authorization Provider

C. Ambari Alert Framework

D. Postgres RDBMS

Question 42
Which hardware feature on an Hadoop datanode is
recommended for cost efficient performance?
Your answer

A. SSD

B. RAID

C. JBOD

D. LVM

Question 43
 Test Blanc
Which data encoding format supports exact storage
of all data in binary representations such as
VARBINARY columns?
Your answer

A. SequenceFiles

B. Flat

C. Parquet

D. RCFile

Question 44
What is the name of the Hadoop-related Apache
project that utilizes an in-memory architecture to run
applications faster than MapReduce?
Your answer

A. Pig

B. Spark

C. Python

D. Hive

Question 45
Which statement about Apache Spark is true?
Your answer

A. It features APIs for C++ and .NET.

 Test Blanc
Your answer

B. It supports HDFS, MS-SQL, and Oracle.

C. It is much faster than MapReduce for complex applications on disk.

D. It runs on Hadoop clusters with RAM drives configured on each DataNode.

Question 46
Which description characterizes a function provided
by Apache Ambari?
Your answer

A. Moves information to/from structured databases.

B. A wizard for installing Hadoop services on host servers.

C. Moves large amounts of streaming event data.

D. A messaging system for real-time data pipelines.

Question 47
What is the preferred replacement for Flume?
Your answer

A. Druid

B. Hortonworks Data Flow

C. Storm

D. NiFi
 Test Blanc
Question 48
What are two primary limitations of MapReduce v1?
Your answer

A. Scalability

B. TaskTrackers can be a bottleneck to MapReduce jobs

C. Resource utilization

D. Number of TaskTrackers limited to 1,000

E. Workloads limited to MapReduce

Question 49
What are two services provided by ZooKeeper?
Your answer

A. Maintaining configuration information.

B. Providing distributed synchronization.

C. Authenticating and auditing user access.

D. Loading bulk data into an Hadoop cluster

Question 50
What are three IBM value-add components to the
Hortonworks Data Platform (HDP)?
Your answer

A. Big Match
 Test Blanc
Your answer

B. Big YARN

C. Big SQL

D. Big Replicate

E. Big Data

F. Big Index

Question 51
What does the user interface for Jupyter look like to
a user?
Your answer

A. Database interface.

B. Linux SSH session.

C. App in web browser.

D. Common desktop app.

Question 52
What is the first step in a data science pipeline?
Your answer

A. Analytics

B. Manipulation
 Test Blanc
Your answer

C. Exploration

D. Acquisition

Question 53
What command is used to list the "magic"
commands in Jupyter?
Your answer

A. %dirmagic

B. %list-all-magic

C. %list-magic

D. %lsmagic

Question 54
What is a "magic" command used for in Jupyter?
Your answer

A. Autoconfiguring data connections using a registry.

B. Parsing and loading data into a notebook.

C. Extending the core language with shortcuts.

D. Running common statistical analyses.

Question 55
Which is an advantage that Zeppelin holds over
Jupyter?
 Test Blanc
Your answer

A. Zeppelin is able to use the R language.

B. Notebooks can be used by multiple people at the same time.

C. Notebooks can be connected to big data engines such as Spark.

D. Users must authenticate before using a notebook.

Question 56
Where does the unstructured data of a project
reside in Watson Studio?
Your answer

A. Database

B. Wrapper

C. Tables

D. Object Storage

Question 57
Before you create a Jupyter notebook in Watson
Studio, which two items are necessary?
Your answer

A. Scala

B. URL

C. Project
 Test Blanc
Your answer

D. Spark Instance

E. File

Question 58
Which type of cell can be used to document and
comment on a process in a Jupyter notebook?
Your answer

A. Kernel

B. Output

C. Code

D. Markdown

Question 59
Which Watson Studio offering used to be available
through something known as IBM Bluemix?
Your answer

A. Watson Studio Cloud

B. Watson Studio Desktop

C. Watson Studio Local

D. Watson Studio Business

Question 60
 Test Blanc
What is the architecture of Watson Studio centered
on?
Your answer

A. Collaborators

B. Data Assets

C. Projects

D. Analytic Assets
 Big Data Engineer v2
IBM Certification
2018

1/ What are the 4Vs of Big Data? (Please select the FOUR that apply)
• Veracity
• Velocity
• Variety
• Volume

2/ What are the three types of Big Data? (Please select the THREE that apply)
• Semi-structured
• Structured
• Unstructured

3/ Select all the components of HDP which provides data access capabilities
• Pig
• MapReduce
• Hive

4/ Select the components that provides the capability to move data from
relational database into Hadoop.
• Sqoop
• Kafka
• Flume

5/ Managing Hadoop clusters can be accomplished using which component?

• Ambari

6/ True or False: The following components are value-add from IBM: Big
Replicate, Big SQL, BigIntegrate, BigQuality, Big Match
• TRUE

7/ True or False: Data Science capabilities can be achieved using only HDP.
FALSE (Big Data Ecosystem UNIT 2)
p.45 // Hortonworks Data Platform.
8/ True or False: Ambari is backed by RESTful APIs for developers to easily
integrate with their own applications.
• True

9/ Which Hadoop functionalities does Ambari provide?

• Manage
• Provision
• Integrate
• Monitor

10/ Which page from the Ambari UI allows you to check the versions of the
software installed on your cluster?
• The Admin > Manage Ambari page

11/ True or False?Creating users through the Ambari UI will also create the
user on the HDFS.
• FALSE
12/ True or False? You can use the CURL commands to issue commands to
Ambari.
• TRUE

13/ True or False: Hadoop systems are designed for transaction processing.
• FALSE

14/ What is the default number of replicas in a Hadoop system?

• 3

15/ True or False: One of the driving principal of Hadoop is that the data is
brought to the program.
FALSE (Big Data Ecosystem UNIT 4)
⇒ programs are brought to the data, not the data to the program

16/ True or False: Atleast 2 Name Nodes are required for a standalone
Hadoop cluster.
FALSE (Big Data Ecosystem UNIT 4)
⇒ One Name Node is required

17/ True or False: The phases in a MR job are Map, Shuffle, Reduce and
Combiner
• TRUE
18/ Centralized handling of job control flow is one of the the limitations of MR
v1.
• TRUE

19/ The Job Tracker in MR1 is replaced by which component(s) in YARN?

• ApplicationMaster
• ResourceManager

20/ What are the benefits of using Spark? (Please select the THREE that
apply)
• Generality
• Speed
• East of use
21/ What are the languages supported by Spark? (Please select the THREE
that apply)
• Python
• Java
• Scala

22/ Resilient Distributed Dataset (RDD) is the primary abstraction of Spark.

• True

23/ What would you need to do in a Spark application that you would not
need to do in a Spark shell to start using Spark?
• Import the necessary libraries to load the SparkContext

24/ True or False: NoSQL database is designed for those that do not want to
use SQL.
FALSE (Big Data Ecosystem UNIT 7)
⇒ Note: HBase and other NoSQL distributed data stores are subject to the CAP
Theorem which states that distributed NoSQL data stores can only achieve 2
out of the 3 properties: consistency, availability and partition tolerance.
25/ Which database is a columnar storage database?
• Hbase

26/ Which database provides a SQL for Hadoop interface?

• Hive

27/ Which Apache project provides coordination of resources?

 • ZooKeeper

28/ What is ZooKeeper's role in the Hadoop infrastructure?

• Manage the coordination between HBase servers
• Hadoop and MapReduce uses ZooKeeper to aid in high
availability of Resource Manager
• Flume uses ZooKeeper for configuration purposes in recent
releases

29/ True or False: Slider provides an intuitive UI which allows you to

dynamically allocate YARN resources.
FALSE (Big Data Ecosystem UNIT 8)
⇒ Apache Slider works in conjunction with YARN to deploy distributed
applications and to monitor them.
30/ True or False: Knox can provide all the security you need within your
Hadoop infrastructure.
FALSE (Big Data Ecosystem UNIT 8)
⇒ Apache Knox provides peripheral security services to an Hadoop
cluster

31/ True or False: Sqoop is used to transfer data between Hadoop and
relational databases.
• True

32/True or False: For Sqoop to connect to a relational database, the JDBC

JAR files for that database must be located in $SQOOP_HOME/bin.
FALSE (Big Data Ecosystem UNIT 9)
⇒ Must copy the JDBC driver JAR files for any relational databases to
$SQOOP_HOME/lib
33/ True or False: Each Flume node receives data as "source", stores it in a
"channel", and sends it via a "sink".
• True

34/ Through what HDP component are Kerberos, Knox, and Ranger managed?
• Ambari

35/ Which security component is used to provide peripheral security?

• Apache Knox

36/ One of the governance issue that Hortonworks DataPlane Service (DPS)
address is visibility over all of an organization's data across all of their
environments — on-prem, cloud, hybrid — while making it easy to maintain
consistent security and governance
• True

37/ True or false: The typical sources of streaming data are Sensors, "Data
exhaust" and high-rate transaction data.
• True

38/ What are the components of Hortonworks Data Flow(HDF)?

• Flow management
• Stream processing
• Enterprise services

39/ True or False: NiFi is a disk-based, microbatch ETL tool that provides flow
management
• True

40/ True or False: MiNiFi is a complementary data collection tool that feeds
collected data to NiFi
• True

41/ What main features does IBM Streams provide as a Streaming Data
Platform? (Please select the THREE that apply)
• Analysis and visualization
• Rich data connections
• Development support
42/ What are the most important computer languages for Data Analytics?
(Please select the THREE that apply)
• Python
• R
• Scala

43/ True or False: GPUs are special-purpose processors that traditionally can
be used to power graphical displays, but for Data Analytics lend themselves to
faster algorithm execution because of the large number of independent
processing cores.
• True

44/ True or False: Jupyter stores its workbooks in files with the .ipynb suffix.
These files can not be stored locally or on a hub server.
FALSE (Introduction to Data Science UNIT 1)
 Course 3

45/ $BIGSQL_HOME/bin/bigsql start

command is used to start Big SQL from the command line?
• True
46/ What are the two ways you can work with Big SQL.
(Please select the TWO that apply)
• Jsqsh
• Web tooling from DSM

47/ What is one of the reasons to use Big SQL?

• Want to access your Hadoop data without using MapReduce
48/ Should you use the default STRING data type?
No (Big SQL UNIT 2)
49/ The BOOLEAN type is defined as SMALLINT SQL type in Big SQL.
• True

50/ Using the LOAD operation is the recommended method for getting data
into your Big SQL table for best performance.
• True

51/ Which file storage format has the highest performance?

• Parquet

52/ What are the two ways to classify functions?

• Built-in functions
• User-defined functions
53/ True or False: UMASK is used to determine permissions on directories
and files.
• True

54/ True or False: You can only Kerberize a Big SQL server before it is
installed. False (Big SQL UNIT 4)
55/ True or False: Authentication with Big SQL only occurs at the Big SQL
layer or the client's application layer. False (Big SQL UNIT 4)
56/ True or False: Ranger and impersonation works well together.
False (Big SQL UNIT 4)
57/ True or False: RCAC can hide rows and columns.
• True

58/ True or False: Nicknames can be used for wrappers and servers.
False (Big SQL UNIT 5)
⇒ Nicknames are created to associate the tables in the remote data
source.

59/ True or False: Server objects defines the property and values of the
connection.
• True

60/ True or False: The purpose of a wrapper provide a library of routines that
doesn't communicates with the data source.
False (Big SQL UNIT 5)

61/ True or False: User mappings are used to authenticate to the remote data
source.
• True
Course 4

62/ True or False: Collaboration with Watson Studio is an optional add-on

component that must be purchased.
False (Watson Studio UNIT 1)

63/ True or False: Watson Studio is designed only for Data Scientists, other
personas would not know how to use it.
False (Watson Studio UNIT 1)
⇒ Thecollaborative platform allows the users, whether they are data
scientists, data engineers, or application developers

64/ True or False: Community provides access to articles, tutorials, and even
data sets that you can use.
• True

65/ True or False: You can import visualization libraries into Watson Studio.
• True

66/ True or False: Collaborators can be given certain access levels.

• True

67/True or False: Watson Studio contains Zeppelin as a notebook

interface.False (Watson Studio UNIT 2) ⇒ Jupyter as notebook
False (Watson Studio UNIT 2) ⇒ Jupyter as notebook Jupyter as notebook