M233: Getting Started with Spark and MongoDB answers

gistfile1.txt

Run a count of the number of documents in the nasa.eva namespace by running: $ mongo > use nasa > db.eva.count() What is the result?
375

Spark guarantees that your processing tasks completes in subseconds
false

Spark is a great system for performing updates on individual records efficiently
false

Spark is a great system for real-time OLTP workloads
false

Level 1
An HDFS deployment can be composed of multiple nodes, each containing a subset of the total data set.
true

Level 2
What does the R and N stand for in YARN?
Resource Negotiator

Level 3
Filtering and projections are functions you would see in which phase of a MapReduce job?
Map phase

Level 4
True or false: I can process data stored in HDFS and write the results of that processing in MongoDB, or can use MongoDB in place of HDFS
True

RDD operations are one of two kinds, actions and transactions
True

The set of operations performed on an RDD is known as the RDD's
lineage

The spark-shell binary is located in which directory?
bin

Which command line parameter is used to identify the spark connector jar?
--conf

Based on the following --conf parameters, what namespace will this Spark process read from. Which namespace will it write to?
Read from profiles.users, write to logs.events

True or False: Since we need this transformation split crew field into on or more tuples we need to use the RDDs flatmap fuction
True

The reduceByKey Method is summing values by astronaut name. What is the unit of that value?
minutes

The spark connector's write target can be set in the same three ways as the read configuration. Two are at the prompt (programatically) and in the spark defaults conf file. What is the third?
command line parameter