About groovy-hadoop

This project’s intention is to bring the power and expressiveness of Groovy to the almighty Hadoop framework to support ad-hoc Hadoop job execution right from the command line with expressive Java or Groovy syntax and without the need to create Java projects and executable JAR files.

Requirements

The only requirement is a working Hadoop installation (the hadoop jar application must be executable). groovy-hadoop was tested with clusters featuring:

• Hadoop-0.20.2 (vanilla Hadoop from Apache)
• Hadoop-0.20.2-cdh3u1-SNAPSHOT (from Cloudera)

Target audience:

• People who like to think and to code in terms of map/reduce but are a little annoyed by the need of setting up Java projects and creating executable Hadoop JAR files for even the most basic map/reduce applications
• People who would like to combine hadoop-streaming with Groovy but do not have the possibility to install Groovy on the Hadoop cluster nodes
• People who would like to run ad-hoc processing jobs and want to use their custom Writable implementations via the Java/Groovy API
• People wanting to prototype map/reduce applications before developing proper Java projects

Who will not benefit from using groovy-hadoop?

• Performance hunters. After all, it is Groovy, it is dynamic, it is not completely pre-compiled. However, it seems to be faster than hadoop-streaming.
• People striving for yet another abstraction layer (TM) on top of map/reduce. Please consider using Pig or Hive or something similar this case.
• Scripting Gurus. If you know how to write your custom queries using your scripting language of choice in concert with hadoop-streaming, chances are that you can do without this project.

License

This project is licensed under the Apache License, Version 2.0. A copy of it can be found in the project’s main folder. In short: this project is (obviously) open source and free.