2. Contents
• Hadoop quick introduction
• An introduction to spark
• Spark – Architecture & Programming Model
1
3. Hadoop
• An Open-Source software for distributed storage of large
dataset on commodity hardware
• Provides a programming model/framework for processing
large dataset in parallel
2
Map
Map
Map
Reduce
Reduce
Input Output
4. Limitations of Map Reduce
• Slow due to replication, serialization, and disk IO
• Inefficient for:
– Iterative algorithms (Machine Learning, Graphs & Network Analysis)
– Interactive Data Mining (R, Excel, Ad hoc Reporting, Searching)
3
Input iter. 1 iter. 2 . . .
HDFS
read
HDFS
write
HDFS
read
HDFS
write
Map
Map
Map
Reduce
Reduce
Input Output
6. Apache Spark
• A big data analytics cluster-computing framework
written in Scala.
• Open Sourced originally in AMPLab at UC Berkley
• Provides in-memory analytics based on RDD
• Highly compatible with Hadoop Storage API
– Can run on top of an Hadoop cluster
• Developer can write programs using multiple
programming languages
5
9. Spark
8
iter. 1 iter. 2 . . .
Input
Not tied to 2 stage Map
Reduce paradigm
1. Extract a working set
2. Cache it
3. Query it repeatedly
Logistic regression in Hadoop and Spark
HDFS
read
11. Spark Programming Model
10
User
(Developer)
Writes
sc=new SparkContext
rDD=sc.textfile(“hdfs://…”)
rDD.filter(…)
rDD.Cache
rDD.Count
rDD.map
Driver Program
RDD
(Resilient
Distributed
Dataset)
• Immutable Data structure
• In-memory (explicitly)
• Fault Tolerant
• Parallel Data Structure
• Controlled partitioning to
optimize data placement
• Can be manipulated using
rich set of operators.
12. RDD
• Programming Interface: Programmer can perform 3
types of operations
11
Transformations
• Create a new dataset
from and existing one.
• Lazy in nature. They
are executed only
when some action is
performed.
• Example :
• Map(func)
• Filter(func)
• Distinct()
Actions
• Returns to the driver
program a value or
exports data to a
storage system after
performing a
computation.
• Example:
• Count()
• Reduce(funct)
• Collect
• Take()
Persistence
• For caching datasets
in-memory for future
operations.
• Option to store on disk
or RAM or mixed
(Storage Level).
• Example:
• Persist()
• Cache()
13. How Spark works
• RDD: Parallel collection with partitions
• User application create RDDs, transform them, and
run actions.
• This results in a DAG (Directed Acyclic Graph) of
operators.
• DAG is compiled into stages
• Each stage is executed as a series of Task (one Task
for each Partition).
12
21. Stage Execution
• Create a task for each Partition in the new RDD
• Serialize the Task
• Schedule and ship Tasks to Slaves
And all this happens internally (you need to do anything)
20
Task 1
Task 2
Task 2
Task 2
23. Summary of Components
• Task: The fundamental unit of execution in Spark
• Stage : Set of Tasks that run parallel
• DAG: Logical Graph of RDD operations
• RDD: Parallel dataset with partitions
22
24. Start the docker container
From
•https://github.com/sequenceiq/docker-spark
docker run -i -t -h sandbox sequenceiq/spark:1.1.1-ubuntu
/etc/ bootstrap.sh –bash
•Run the spark shell using yarn or local
spark-shell --master yarn-client --driver-memory 1g --executor-memory
1g --executor-cores 2
23
25. Running the example and Shell
• To Run the examples
– $ run-example SparkPi 10
• We can start a spark shell via
– spark-shell -- master local n
• The -- master specifies the master URL for a
distributed cluster
• Example applications are also provided in Python
– spark-submit example/src/main/python/pi.py 10
24
26. Collections and External Datasets
• A Collection can be parallelized using the SparkContext
– val data = Array(1, 2, 3, 4, 5)
– val distData = sc.parallelize(data)
• Spark can create distributed dataset from HDFS, Cassandra,
Hbase, Amazon S3, etc.
• Spark supports text files, Sequence Files and any other
Hadoop input format
• Files can be read from an URI local or remote (hdfs://, s3n://)
– scala> val distFile = sc.textFile("data.txt")
– distFile: RDD[String] = MappedRDD@1d4cee08
– distFile.map(s => s.length).reduce((a,b) => a + b)
25
27. RDD operations
• Count the length of the words in the file
– val lines = sc.textFile("data.txt")
– val lineLengths = lines.map(s => s.length)
– val totalLength = lineLengths.reduce((a, b) => a + b)
• If we want to use lineLengths later we can run
– lineLengths.persist()
• This will save in the memory the value of lineLengths
before reducing
26
28. Passing a function to Spark
• Spark is based on Anonymous function syntax
– (x: Int) => x *x
• Which is a shorthand for
new Function1[Int,Int] {
def apply(x: Int) = x * x
}
• We can define functions with more parameters and without
– (x: Int, y: Int) => "(" + x + ", " + y + ")”
– () => { System.getProperty("user.dir") }
• The syntax is a shorthand for
– Funtion1[T,+E] … Function22[…]
27
29. Passing a function to Spark
object MyFunctions {
def func1(s: String): String = s + s
}
file.map(MyFunctions.func1)
class MyClass {
def func1(s: String): String = { ... }
def doStuff(rdd: RDD[String]): RDD[String] = { rdd.map(func1) }
}
28
30. Working with Key-Value Pairs
• We can setup RDD with key-value pairs that are
caster to Tuple2 type
– val lines = sc.textFile("data.txt")
– val pairs = lines.map(s => (s, 1))
– val counts = pairs.reduceByKey((a, b) => a + b)
• We can use counts.sortByKey() to sort
• And finally counts.collect() to bring them back
• NOTE: when using custom objects as key-value we
should be sure that they have the method equals()
with hashcode()
http://docs.oracle.com/javase/7/docs/api/java/lang/Object.ht
ml#hashCode()
29
31. Transformations
• There are several transformations supported by
Spark
– Map
– Filter
– flatMap
– mapPartitions
– ….
• When they are executed?
30
32. Actions
• The following table lists some of the common actions
supported:
– Reduce
– Collect
– Count
– First
– Take
– takeSample
31
33. RDD Persistence
• One of the most important capabilities in Spark is persisting
(or caching) a dataset in memory across operations
• Caching is a key tool for iterative algorithms and fast
interactive use
• You can mark an RDD to be persisted using the persist() or
cache() methods on it
• The first time it is computed in an action, it will be kept in
memory on the nodes. Spark’s cache is fault-tolerant – if any
partition of an RDD is lost, it will automatically be recomputed
using the transformations that originally created it.
32
34. RDD persistence
• In addition, each persisted RDD can be stored using a different
storage level,
• for example we can persist
– the dataset on disk,
– in memory but as serialized Java objects (to save space), replicate it
across nodes,
– off-heap in Tachyon
• Note: In Python, stored objects will always be serialized with
the Pickle library, so it does not matter whether you choose a
serialized level.
• Spark also automatically persists some intermediate data in
shuffle operations (e.g. reduceByKey), even without users
calling persist
33
35. Which Storage Level to Choose?
• Memory only if that fit in the main memory
• If not, try using MEMORY_ONLY_SER and selecting a fast
serialization library to make the objects much more space-
efficient, but still reasonably fast to access.
• Don’t spill to disk unless the functions that computed your
datasets are expensive, or they filter a large amount of the
data. Otherwise, recomputing a partition may be as fast as
reading it from disk.
• Use the replicated storage levels if you want fast fault
recovery
• Use OFF_HEAP in environments with hig amounts of memory
used or multiple applications
34
36. Shared Variables
• Normally when functions are executed on a remote
node it works on immutable copies
• However, Sparks does provide two types of shared
variables for two usages:
– Broadcast variables
– Accumulators
35
37. Broadcast Variables
• Broadcast variables allow the programmer to keep a
read-only variable cached on each machine rather
than shipping a copy of it with tasks.
scala> val broadcastVar = sc.broadcast(Array(1, 2, 3))
broadcastVar: org.apache.spark.broadcast.Broadcast[Array[Int]] =
Broadcast(0)
scala> broadcastVar.value
res0: Array[Int] = Array(1, 2, 3)
36
38. Accumulators
• Accumulators are variables that are only “added” to through
an associative operation and can therefore be efficiently
supported in parallel
• Spark natively supports accumulators of numeric types, and
programmers can add support for new types
• Note: not yet supported on Python
scala> val accum = sc.accumulator(0, "My Accumulator")
accum: spark.Accumulator[Int] = 0
scala> sc.parallelize(Array(1, 2, 3, 4)).foreach(x => accum += x)
scala> accum.value
res7: Int = 10
37
40. Spark Examples
• Let’s walk through
http://spark.apache.org/examples.html
• Other examples are on
• Basic Sample
=>https://github.com/apache/spark/tree/master/exa
mples/src/main/scala/org/apache/spark/examples
• Streaming Samples =>
https://github.com/apache/spark/tree/master/exam
ples/src/main/scala/org/apache/spark/examples/stre
aming
39
41. Create a Self Contained App in
Scala
/* SimpleApp.scala */
import org.apache.spark.SparkContext
import org.apache.spark.SparkContext._
import org.apache.spark.SparkConf
object SimpleApp {
def main(args: Array[String]) {
val logFile = "YOUR_SPARK_HOME/README.md" // Should be some file on your system
val conf = new SparkConf().setAppName("Simple Application")
val sc = new SparkContext(conf)
val logData = sc.textFile(logFile, 2).cache()
val numAs = logData.filter(line => line.contains("a")).count()
val numBs = logData.filter(line => line.contains("b")).count()
println("Lines with a: %s, Lines with b: %s".format(numAs, numBs))
}
}
40
42. Create a Self Contained App in
Scala
Create a build.sbt file
name := "Simple Project"
version := "1.0"
scalaVersion := "2.10.4"
libraryDependencies += "org.apache.spark" %% "spark-core" % "1.2.0"
41
43. Project folder
• That how the project directory should look
$ find .
.
./simple.sbt
./src
./src/main
./src/main/scala
./src/main/scala/SimpleApp.scala
• With sbt package we can create the jar
• To submit the job
$ YOUR_SPARK_HOME/bin/spark-submit
--class "SimpleApp"
--master local[4]
target/scala-2.10/simple-project_2.10-1.0.jar
42
46. A simple example
• We create a local StreamingContext with two execution
threads, and batch interval of 1 second.
import org.apache.spark._
import org.apache.spark.streaming._
import org.apache.spark.streaming.StreamingContext._
// Create a local StreamingContext with two working thread and batch
interval of 1 second.
// The master requires 2 cores to prevent from a starvation scenario.
val conf = new
SparkConf().setMaster("local[2]").setAppName("NetworkWordCount")
val ssc = new StreamingContext(conf, Seconds(1))
45
47. A sample example
• Using this context, we can create a DStream that represents
streaming data from a TCP source
val lines = ssc.socketTextStream("localhost", 9999)
• Split each line into words
val words = lines.flatMap(_.split(" "))
• Count each word in the batch
import org.apache.spark.streaming.StreamingContext._
val pairs = words.map(word => (word, 1))
val wordCounts = pairs.reduceByKey(_ + _)
wordCounts.print()
46
48. A sample example
• Note that when these lines are executed, Spark Streaming
only sets up the computation it will perform when it is
started, and no real processing has started yet
ssc.start() // Start the computation
ssc.awaitTermination() // Wait for the computation to terminate
• Start netcat as data server by using
– Nc –lk 9999
47
49. A sample example
• If you have already downloaded and built Spark, you
can run this example as follows. You will first need to
run Netcat (a small utility found in most Unix-like
systems) as a data server by using
– nc -lk 9999
• Run the example by
– run-example streaming.NetworkWordCount localhost 9999
• http://spark.apache.org/docs/latest/streaming-
programming-guide.html
48
Editor's Notes
Resilient Distributed Datasets or RDD are the distributed memory abstractions that lets programmer perform in-memory parallel computations on large clusters. And that too in a highly fault tolerant manner.
This is the main concept around which the whole Spark framework revolves around.
Currently 2 types of RDDs:
Parallelized collections: Created by calling parallelize method on an existing Scala collection. Developer can specify the number of slices to cut the dataset into. Ideally 2-3 slices per CPU.
Hadoop Datasets: These distributed datasets are created from any file stored on HDFS or other storage systems supported by Hadoop (S3, Hbase etc). These are created using SparkContext’s textFile method. Default number of slices in this case is 1 slice per file block.
Transformations: Like map – takes an RDD as an input, passes & process each element to a function, and return a new transformed RDD as an output.
By default, each transformed RDD is recomputed each time you run an action on it. Unless you specify the RDD to be cached in memory. Spark will try to keep the elements around the cluster for faster access.
RDD can be persisted on discs as well.
Caching is the Key tool for iterative algorithms.
Using persist, one can specify the Storage Level for persisting an RDD. Cache is just a short hand for default storage level. Which is MEMORY_ONLY.
MEMORY_ONLY
Store RDD as deserialized Java objects in the JVM. If the RDD does not fit in memory, some partitions will not be cached and will be recomputed on the fly each time they're needed. This is the default level.
MEMORY_AND_DISK
Store RDD as deserialized Java objects in the JVM. If the RDD does not fit in memory, store the partitions that don't fit on disk, and read them from there when they're needed.
MEMORY_ONLY_SER
Store RDD as serialized Java objects (one byte array per partition). This is generally more space-efficient than deserialized objects, especially when using a fast serializer, but more CPU-intensive to read.
MEMORY_AND_DISK_SER
Similar to MEMORY_ONLY_SER, but spill partitions that don't fit in memory to disk instead of recomputing them on the fly each time they're needed.
DISK_ONLY
Store the RDD partitions only on disk.
MEMORY_ONLY_2, MEMORY_AND_DISK_2 etc
Same as the levels above, but replicate each partition on two cluster nodes.
Which Storage level is best:
Few things to consider:
Try to keep in-memory as much as possible
Try not to spill to disc unless your computed datasets are memory expensive
Use replication only if you want fault tolerance
