As this project is for learning purposes, we have created separate class/object for each use case with main function.

Here are the details about using this project.

• Clone the repository.
• Open with IntelliJ or Eclipse with Scala and sbt plugin.
• Go to the respective program and identify the arguments.
• Validate either using IDE or by using local spark or based up on the instructions provided.

Let us understand how to setup the project.

• Clone GitHub Repository.

git clone https://github.com/dgadiraju/spark2demo.git

• It will create a folder by name spark2demo
• We are externalizing properties such as input directory, output directory etc.
• Before running any program or building jar file, make sure to go to src/main/resources/application.properties and do the following:
  • Review and modify the properties as per your paths.
  • Make sure the paths defined in application.properties are correct.
  • Make sure file formats used in the programs and files in the defined paths are consistent.
• Go to spark2demo and run sbt package
• It will build the jar file in target/scala-2.11 folder.

We will be having several programs as part of this repository.

• Each one of it will have a main function.
• We will see how each program can run locally or on ITVersity labs or on EMR Cluster.

Let us understand how to run GetDailyProductRevenue using different options.

• Validate using IDE
• Run using Local Spark
• Run on ITVersity labs
• Run on EMR Cluster.

Here is how you can validate the program locally.

• Go to the program you want to run - GetDailyProductRevenue
• Right click and then click on Run 'GetDailyProductRevenue'
• Program might fail for the first time as we have to pass arguments to it.
• If it fails - go to Run -> Edit Configurations and pass program arguments.
• This program takes only one argument and we need to pass dev to validate locally.
• If successful the program will exit with exit code 0.
• Make sure to validate by going to the output directory.

Let us ensure that jar file is built so that we can submit the job using Spark that is setup locally using local mode.

• Go to working directory and run sbt package
• It will create jar file in this location - target/scala-2.11/spark2demo_2.11-0.1.jar
• As the application is dependent on 3rd party plugin called as typesafe config, we need to pass the jar file while running job using spark-submit.
• We can pass it either by using --packages com.typesafe:config:1.3.2 or by using --jars ~/.ivy2/jars/com.typesafe_config-1.3.2.jar.
• .ivy2 is the directory that is created as part of setting up sbt based application. It is used to cache the jar files that are required for sbt based applications.
• spark-submit command looks like this to run the spark application using local spark in local mode.

spark-submit \
  --class retail_db.GetDailyProductRevenue \
  --packages com.typesafe:config:1.3.2 \
  target/scala-2.11/spark2demo_2.11-0.1.jar dev