GPU621/Apache Spark Fall 2022

Apache Spark Core API

RDD Overview

One of the most important concepts in Spark is a resilient distributed dataset (RDD). RDD is a collection of elements partitioned across the nodes of the cluster that can be operated in parallel. RDDs are created by starting with a file, or an existing Java collection in the driver program, and transforming it. We will introduce some key APIs provided by Spark Core 2.2.1 using Java 8. You can find more information about the RDD here. https://spark.apache.org/docs/2.2.1/rdd-programming-guide.html

Spark Library Installation Using Maven

An Apache Spark application can be easily instantiated using Maven. To add the required libraries, you can copy and paste the following code into the "pom.xml".

   <properties>
       <maven.compiler.source>8</maven.compiler.source>
       <maven.compiler.target>8</maven.compiler.target>
   </properties>
   <dependencies>
       <dependency>
           <groupId>org.apache.spark</groupId>
           <artifactId>spark-core_2.10</artifactId>
           <version>2.2.0</version>
       </dependency>
       <dependency>
           <groupId>org.apache.spark</groupId>
           <artifactId>spark-sql_2.10</artifactId>
           <version>2.2.0</version>
       </dependency>
       <dependency>
           <groupId>org.apache.hadoop</groupId>
           <artifactId>hadoop-hdfs</artifactId>
           <version>2.2.0</version>
       </dependency>
   </dependencies>

Create And Set Up Spark

Spark needs to be set up in a cluster so first we need to create a JavaSparkContext object, which tells Spark how to access a cluster. To create a SparkContext you first need to build a SparkConf object that contains information about your application. We will talk about how to set up a spark in a cluster later. Now let's try to create a spark locally. To do that, we will need the following code:

  //create and set up spark
  SparkConf conf = new SparkConf().setAppName("HelloSpark").setMaster("local[*]");
  JavaSparkContext sc = new JavaSparkContext(conf);
  sc.setLogLevel("WARN");

Create RDDs

There are two ways to create RDDs: parallelizing an existing collection in your driver program, or referencing a dataset in an external storage system, such as a shared filesystem, HDFS, HBase, or any data source offering a Hadoop InputFormat.

1. Parallelized Collections Let’s start with some Java collections by calling JavaSparkContext’s parallelize method on an existing Collection in your driver program. The elements of the collection are copied to form a distributed dataset that can be operated on in parallel.

       //create input data list
       List<Integer> inputData = new ArrayList<>();
       inputData.add(11);
       inputData.add(22);
       inputData.add(33);
       inputData.add(44);

       //use RDD to run create RDDS
       JavaRDD<Integer> javaRDD = sc.parallelize(inputData);

2. External Datasets The other way is to create RDD from any storage source supported by Hadoop, including your local file system, HDFS, Amazon S3, etc. Text file RDDs can be created using SparkContext’s textFile method. This method takes an URI for the file (either a local path on the machine, or a hdfs://, s3n://, etc URI) and reads it as a collection of lines.

       //From local file
       JavaRDD<String> sentences = sc.textFile("src/main/resources/subtitles/input.txt");
       //From a S3 file
       JavaRDD<String> sentences = sc.textFile("s3://gpu621-demo/input.txt");

Deploy Apache Spark Application On AWS

Amazon EMR is a cloud big data solution for petabyte-scale data processing, interactive analytics, and machine learning using open-source frameworks such as Apache Spark, Apache Hive, and Presto provided by AWS cloud service. EMR is easy to use and it has low cost, so it’s a great start for spark beginners.

Prerequisite

From here, I will assume you have an AWS service account and that you have basic knowledge about AWS services like how to use S3 bucket, or how to add role or policy to services. Also, you will need to have basic knowledge about SSH and Linux commands.