Intrdocution

This is a very very simple mapreduce application. You can run this application in command with following command.

./gradlew run    

This command will download all the jars to your local machine. No hadoop installation package is required, no setting is required, just issue the command and does work!!

The output would print this information

The output goes to: out/20140520170354

Just use

cat out/20140520170354/part-r-00000

To see the result.

Eclipse Integration

Please follow the steps below

1. Issue this command to create the eclise project files

   ./gradlew eclipse

2. Open the eclipse. Select the "Import" > "Existing Projects into Workspace".

3. Run as application

Run in hadoop cluster

I assume that your have knowledge to set up the hadoop standalone server or cluster. Please follow the steps below

1. Issue this command to create the executable jar

   ./gradlew jar

2. Run in hadoop

   hadoop jar build/libs/wordcount.jar

Good luck and happy MapReduce!!!

Azure HDInsigh Support

To make this example more-Hadoop-like, and make it run more smoothly under HDInsight, I add a new task 'hadoop', and define an argument named 'args' to pass all the required parameters. ( Thank @MonsterSupreme )