stud
bud
Sign up
Log in
Home
Library
Explore
SAT
More
Game 7 - Spark Flashcards | StudBud
Explore
/
Game 7 - Spark
Game 7 - Spark
30 terms
·
StudBud Administration
·
1 play
Let this quiz inspire your own quiz for formative assessment.
Flashcards
30 cards
F
Practice quiz
P
Host live
L
Questions
30
Answers
1
Which of the following is not considered a best practice in Spark?
Use take in production instead of collect
Cache commonly used dataframes
Use python codes for processing data wherever possible
Apply transformations and actions wherever possible
2
What is the main purpose of Spark?
On disk processing of big data leveraging MapReduce
On disk processing of big data leveraging HiveQL
In memory distributed processing of big data
In-memory real-time processing of big data
3
What command do we apply to create a dataframe from a collection?
sqlContext.createDataFrame
sqlContext.createRDD
sqlContext.Parallelize
sqlContext.createData
4
What is the purpose of the SQL Context?
Tells Spark how to implement MapReduce operations
Sets up the PySpark environment
Tells Spark how and where to access a cluster
Allows Spark to create Spark dataframes
5
Which of the following is the underlying execution engine for Spark?
Spark SQL
MLlib
Spark Core
Spark Streaming
6
What is the advantage of using Spark over MapReduce?
Spark leverages HDFS as a data store
RDD is flexible and modifiable
RDD supports in-memory distributed processing
RDD eliminates the need for replication factor in HDFS
7
Which of the following uses Spark Core's fast scheduling capability to perform streaming analytics?
Spark Core
Spark SQL
Spark Streaming
GraphX
8
Which of the following commands is not a Spark transformation?
select
filter
sort
collect
9
Which of the following is not a way of creating a RDD?
Referencing a dataset in AWS S3
Modifying an existing RDD
Parallelizing an existing collection in the driver program
Referencing a dataset in HDFS
10
What are the requirements prior to installing Spark?
Java and Scala
Java and Hadoop
Java and Python
a) Python and R
11
Which of the following is not a feature of a RDD?
Can contain Python, Java or Scala objects
Fault-tolerant collection of objects
A fundamental data structure of Spark
Read/Write enabled, partitioned collection of records
12
What command do we apply to create a dataframe from an external text file?
sqlContext.read.file
sqlContext.createDataFrame
sqlContext.read
sqlContext.read.txt
13
What is the purpose of the Spark Context?
Allows Spark to create Spark dataframes
Tells Spark how to implement MapReduce operations
Tells Spark how and where to access a cluster
Sets up the PySpark environment
14
Which of the following commands is not a Spark action?
show
count
take
groupBy
15
Which of the following organizations originally developed Spark?
Bekeley’s AMPLab
Facebook
Google
Cloudera
16
What is a RDD?
An intermediate result of a MapReduce operation
An immutable distributed collection of objects
An intermediate result of a Spark operation
A mutable distributed collection of objects
17
What is the disadvantage of using MapReduce?
Distributed processing paradigm makes data provisioning slow
Replication, serialization and disk IO makes MapReduceslow
HDFS requires high performance machines to set up
Distributed processing paradigm makes it challenging to consolidate data
18
Which of the following is not an advantage in using Spark Transformation?
Recovers from failures and slow workers
Catalyst to optimize required calculations
Lazy evaluation
Replicates partitions for distributed storage
19
How can we construct Spark dataframes?
Transforming existing Spark or Pandas dataframes
Referencing files from HDFS
All of the above
Parallelize existing python collections
20
What does RDD stand for?
Responsive Distributed Dataset
Responsive Distributed Dataset
Resilient Distributed Dataset
Resilient Distributed Dataframe
21
What is the fundamental data concept from Spark 2.0 onwards?
Struct
Dictionary
Map
Dataframe
22
What are the components associated with the Spark Driver Program?
Spark Context and Cluster Manager
Cluster Manager and Worker
Spark Context and SQL Context
Spark Context and Executor
23
What are the requirements prior to installing PySpark?
Python and Spark
Python and Scala
a) Java and Scala
Python and Java
24
Which of the following is not a feature of Spark?
Has its own distributed file system akin to HDFS
Supports SQL queries, stream analysis and machine learning
Up to 100 times faster than Hadoop
Supports multiple programming languages
25
What is a data model?
A machine learning model for pattern recognition
A description of data types associated with a dataset
An analytical model for predictive analysis
A collection of concepts for describing data
26
What are the two broad type of operations in Spark?
Spark Context and Spark SQL
Parallelization and Transformation
Transformation and Activation
Transformation and Action
27
What is the advantage of using Spark dataframe?
Flexible, mutable structure
Enables sequential read-write operations on disk
Enables sequential operations on data collections
Track lineage information for efficient recompute
28
Which of the following is not a component in Spark?
Mahout (Machine Learning)
Spark Core
Spark SQL
Spark Streaming
29
Which of the following is the Spark component comprising the data abstraction known as SchemaRDD?
Spark Streaming
Spark SQL
Spark Core
MLlib
30
What do we mean when say Transformations are lazy?
Slow in execution
Not replicating data blocks for parallel processing
Not computed immediately but consolidated for execution
Not replicating data blocks for distributed storage
