Spark crash course workshop at Hadoop Summit

© Hortonworks Inc. 2014
In-memory processing with Apache Spark
Dhruv Kumar and Saptak Sen
Hortonworks. We do Hadoop.
June 9, 2015

About the presenters
Saptak Sen
Technical Product Manager
Hortonworks Inc.
Dhruv Kumar
Partner Solutions Engineer.
Hortonworks Inc.

In this workshop
• Introduction to HDP and Spark
• Installing Spark on HDP
• Spark Programming
• Core Spark: working with RDDs
• Spark SQL: structured data access
• Conclusion and Further Reading, Q/A

Installing Spark on HDP

Installing Spark on HDP
• If you have Hortonworks Sandbox with HDP 2.2.4.2 you have Spark 1.2.1
• If you have Hortonworks Sandbox with HDP 2.3 Preview you have Spark 1.3.1
• If you have Hortonworks Sandbox on Azure you will need to install Spark
For instructions and workshop content goto http://saptak.in/spark
• GA of Spark 1.3.1
– Fully supported by Hortonworks
– Install with Ambari HDP 2.2.2. Other combination unsupported.

Introduction to HDP and Spark

HDP delivers a comprehensive data management platform
HDP 2.2
Hortonworks Data Platform
Provision,
Manage &
Monitor
Ambari
Zookeeper
Scheduling
Oozie
Data Workflow,
Lifecycle &
Governance
Falcon
Sqoop
Flume
NFS
WebHDFS
YARN: Data Operating System
DATA MANAGEMENT
SECURITY
BATCH, INTERACTIVE & REAL-TIME
DATA ACCESS
GOVERNANCE
& INTEGRATION
Authentication
Authorization
Accounting
Data Protection
Storage: HDFS
Resources: YARN
Access: Hive, …
Pipeline: Falcon
Cluster: Knox
OPERATIONS
Script
Pig
Search
Solr
SQL
Hive
HCatalog
NoSQL
HBase
Accumulo
Stream
Storm
Other
ISVs
1 ° ° ° ° ° ° ° ° °
° ° ° ° ° ° ° ° ° °
° ° ° ° ° ° ° ° ° °
°
°
N
HDFS
(Hadoop Distributed File System)
In-Memory
Spark
Deployment Choice
Linux Windows On-
Premise
Cloud
YARN is the architectural
center of HDP
• Enables batch, interactive
and real-time workloads
• Single SQL engine for both batch
and interactive
• Enables best of breed ISV tools to
deeply integrate into Hadoop via YARN
Provides comprehensive
enterprise capabilities
• Governance
• Security
• Operations
The widest range of
deployment options
• Linux & Windows
• On premise & cloud
TezTez

Let’s drill into one workload … Spark
HDP 2.1
Provision,
Manage &
Monitor
Ambari
Zookeeper
Scheduling
Oozie
Data Workflow,
Lifecycle &
Governance
Falcon
Sqoop
Flume
NFS
WebHDFS
DATA MANAGEMENT
SECURITY
DATA ACCESS
GOVERNANCE
& INTEGRATION
Authentication
Authorization
Accounting
Data Protection
Storage: HDFS
Resources: YARN
Access: Hive, …
Pipeline: Falcon
Cluster: Knox
OPERATIONS
Script
Pig
Search
Solr
SQL
Hive
HCatalog
NoSQL
HBase
Accumulo
Stream
Storm
Other
ISVs
1 ° ° ° ° ° ° ° ° °
° ° ° ° ° ° ° ° ° °
° ° ° ° ° ° ° ° ° °
°
°
N
HDFS
(Hadoop Distributed File System)
Deployment Choice
Linux Windows On-
Premise
Cloud
YARN is the architectural
center of HDP
• Enables batch, interactive
and real-time workloads
• Single SQL engine for both batch
and interactive
• Enables best of breed ISV tools to
deeply integrate into Hadoop via YARN
Provides comprehensive
• Governance
• Security
• Operations
The widest range of
deployment options
• Linux & Windows
• On premise & cloud
TezTez
In-Memory

What is Spark?
• Spark is
– an open-source software solution that performs rapid calculations
on in-memory datasets
- Open Source [Apache hosted & licensed]
• Free to download and use in production
• Developed by a community of developers
- In-memory datasets
• RDD (Resilient Distributed Data) is the basis for what Spark enables
• Resilient – the models can be recreated on the fly from known state
• Distributed – the dataset is often partitioned across multiple nodes for
increased scalability and parallelism

Spark Components
Spark allows you to do data processing, ETL, machine learning,
stream processing, SQL querying from one framework

Why Spark?
• One tool for data engineering and data science tasks
• Native integration with Hive, HDFS and any Hadoop FileSystem
implementation
• Faster development: concise API, Scala (~3x lesser code than Java)
• Faster execution: for iterative jobs because of in-memory caching (not
all workloads are faster in Spark)
• Promotes code reuse: APIs and data types are similar for batch and
streaming

Hortonworks Commitment to Spark
Hortonworks is focused on making
Apache Spark enterprise ready so
you can depend on it for mission
critical applications
SECURITY
DATA ACCESS
GOVERNANCE
&INTEGRATION
OPERATIONS
Script
Pig
Search
Solr
SQL
Hive
HCatalog
NoSQL
HBase
Accumulo
Stream
Storm
Other
ISVs
TezTez
In-Memory
1. YARN enable Spark to
co-exist with other engines
Spark is “YARN Ready” so its
memory & CPU intensive apps
can work with predictable
performance along side other
engines all on the same set(s) of
data.
2. Extend Spark with
Ensure Spark can be managed,
secured and governed all via a
single set of frameworks to
ensure consistency. Ensure
reliability and quality of service of
Spark along side other engines.
3. Actively collaborate within
the open community
As with everything we do at
Hortonworks we work entirely
within the open community
across Spark and all related
projects to improve this key
Hadoop technology.

Reference Deployment Architecture
Batch Source
Streaming
Source
Reference Data
Stream Processing
Storm/Spark-Streaming
Data Pipeline
Hive/Pig/Spark
Long Term Data
Warehouse
Hive + ORC
Data Discovery
Operational
Reporting
Business
Intelligence
Ad Hoc/On
Demand Source
Data Science
Spark-ML, Spark-SQL
Advanced
Analytics
Data Sources Data Processing, Storage & Analytics Data Access

Spark Deployment Modes
Mode setup with
Ambari
• Spark Standalone Cluster
– For developing Spark apps against a local Spark (similar to develop/deploying in IDE)
• Spark on YARN
– Spark driver (SparkContext) in YARN AM(yarn-cluster)
– Spark driver (SparkContext) in local (yarn-client)
• Spark Shell runs in yarn-client only
Client
Executor
App
Master
Client
Executor
App
Master
Spark Driver
Spark Driver
YARN-Client YARN-Cluster

Spark on YARN
YARN RM
App Master
Monitoring UI

Programming Spark

How Does Spark Work?
• RDD
• Your data is loaded in parallel into structured collections
• Actions
• Manipulate the state of the working model by forming new RDDs
and performing calculations upon them
• Persistence
• Long-term storage of an RDD’s state

Example RDD Transformations
•map(func)
•filter(func)
•distinct(func)
• All create a new DataSet from an existing one
• Do not create the DataSet until an action is performed (Lazy)
• Each element in an RDD is passed to the target function and the
result forms a new RDD

Example Action Operations
•count()
•reduce(func)
•collect()
•take()
• Either:
• Returns a value to the driver program
• Exports state to external system

Example Persistence Operations
•persist() -- takes options
•cache() -- only one option: in-memory
• Stores RDD Values
• in memory (what doesn’t fit is recalculated when necessary)
• Replication is an option for in-memory
• to disk
• blended

1. Resilient Distributed Dataset [RDD] Graph
val v = sc.textFile("hdfs://…some-hdfs-data")
mapmap reduceByKey collecttextFile
v.flatMap(line=>line.split(" "))
.map(word=>(word, 1)))
.reduceByKey(_ + _, 3)
.collect()
RDD[String]
RDD[List[String]]
RDD[(String, Int)]
Array[(String, Int)]
RDD[(String, Int)]

Processing A File in Scala
//Load the file:
val file = sc.textFile("hdfs://…/user/DAW/littlelog.csv")
//Trim away any empty rows:
val fltr = file.filter(_.length > 0)
//Print out the remaining rows:
fltr.foreach(println)
22

Looking at the State in the Machine
//run debug command to inspect RDD:
scala> fltr.toDebugString
//simplified output:
res1: String =
FilteredRDD[2] at filter at <console>:14
MappedRDD[1] at textFile at <console>:12
HadoopRDD[0] at textFile at <console>:12
23

A Word on Anonymous Functions
Scala programmers make great use of anonymous functions as can
be seen in the code:
flatMap( line => line.split(" ") )
24
Argument
to the
function
Body of
the
function

Scala Functions Come In a Variety of Styles
flatMap( line => line.split(" ") )
flatMap((line:String) => line.split(" "))
flatMap(_.split(" "))
25
Argument to the
function (type inferred)
Body of the function
Argument to the
function (explicit type)
Body of the
function
No Argument to the
function declared
(placeholder) instead
Body of the function includes placeholder _ which allows for exactly one use of
one arg for each _ present. _ essentially means ‘whatever you pass me’

And Finally – the Formal ‘def’
def myFunc(line:String): Array[String]={
return line.split(",")
}
//and now that it has a name:
myFunc("Hi Mom, I’m home.").foreach(println)
Return type of the function)
Body of the function
Argument to the function)

Things You Can Do With RDDs
• RDDs are objects and expose a rich set of methods:
27
Name Description Name Description
filter Return a new RDD containing only those
elements that satisfy a predicate
collect Return an array containing all the elements of
this RDD
count Return the number of elements in this
RDD
first Return the first element of this RDD
foreach Applies a function to all elements of this
RDD (does not return an RDD)
reduce Reduces the contents of this RDD
subtract Return an RDD without duplicates of
elements found in passed-in RDD
union Return an RDD that is a union of the passed-in
RDD and this one

More Things You Can Do With RDDs
• More stuff you can do…
28
Name Description Name Description
flatMap Return a new RDD by first applying a
function to all elements of this RDD, and
then flattening the results
checkpoint Mark this RDD for checkpointing (its state will
be saved so it need not be recreated from
scratch)
cache Load the RDD into memory (what
doesn’t fit will be calculated as needed)
countByValue Return the count of each unique value in this
RDD as a map of (value, count) pairs
distinct Return a new RDD containing the
distinct elements in this RDD
persist Store the RDD to either memory, Disk, or
hybrid according to passed in value
sample Return a sampled subset of this RDD unpersist Clear any record of the RDD from disk/memory

Code ‘select count’
Equivalent SQL Statement:
Select count(*) from pagecounts WHERE state = ‘FL’
Scala statement:
val file = sc.textFile("hdfs://…/log.txt")
val numFL = file.filter(line =>
line.contains("fl")).count()
scala> println(numFL)
29
1. Load the page as an RDD
2. Filter the lines of the page
eliminating any that do not
contain “fl“
3. Count those lines that
remain
4. Print the value of the
counted lines containing ‘fl’

Spark SQL
30

What About Integration With Hive?
scala> val hiveCTX = new org.apache.spark.sql.hive.HiveContext(sc)
scala> hiveCTX.hql("SHOW TABLES").collect().foreach(println)
…
[omniture]
[omniturelogs]
[orc_table]
[raw_products]
[raw_users]
…
31

More Integration With Hive:
scala> hCTX.hql("DESCRIBE raw_users").collect().foreach(println)
[swid,string,null]
[birth_date,string,null]
[gender_cd,string,null]
scala> hCTX.hql("SELECT * FROM raw_users WHERE gender_cd='F' LIMIT
5").collect().foreach(println)
[0001BDD9-EABF-4D0D-81BD-D9EABFCD0D7D,8-Apr-84,F]
[00071AA7-86D2-4EB9-871A-A786D27EB9BA,7-Feb-88,F]
[00071B7D-31AF-4D85-871B-7D31AFFD852E,22-Oct-64,F]
[000F36E5-9891-4098-9B69-CEE78483B653,24-Mar-85,F]
[00102F3F-061C-4212-9F91-1254F9D6E39F,1-Nov-91,F]
32

Querying RDD Using SQL
// SQL statements can be run directly on RDD’s
val teenagers =
sqlC.sql("SELECT name FROM people
WHERE age >= 13 AND age <= 19")
// The results of SQL queries are SchemaRDDs and support
// normal RDD operations:
val nameList = teenagers.map(t => "Name: " + t(0)).collect()
// Language integrated queries (ala LINQ)
val teenagers =
people.where('age >= 10).where('age <= 19).select('name)

Conclusion and Resources

Conclusion
• Spark is a unified framework for data engineering and data
science
• Spark can be programmed in Scala, Java and Python.
• Spark issupported by Hortonworks
• Certain workloads are faster in Spark because of in-memory
caching.

References and Further Reading
• Apache Spark website: https://spark.apache.org/
• Hortonworks Spark website: http://hortonworks.com/hadoop/spark/
• Hortonworks Sandbox Tutorials https://hortonworks.com/tutorials
• “Learning Spark” by O’Reilly Publishers

Spark crash course workshop at Hadoop Summit

More Related Content

What's hot

Viewers also liked

Similar to Spark crash course workshop at Hadoop Summit

More from DataWorks Summit

Recently uploaded

Spark crash course workshop at Hadoop Summit

Editor's Notes