Jumbune is an open-source product built for analyzing Hadoop cluster and MapReduce jobs. It provides development & administrative insights of Hadoop based analytical solutions. It enables user to Debug, Profile, Monitor & Validate analytical solutions hosted on decoupled clusters.
http://jumbune.org/jira/browse/JUM
Jumbune can be build with maven
- To build for a Yarn Hadoop cluster
mvn clean install -P yarn
- To build for a non Yarn Hadoop cluster
mvn clean install
Detailed installation guide can be found at http://bit.ly/1kBG4Qo
Deployment Planning guide at http://bit.ly/1oiXGk2
- Deploying Jumbune
java -jar <location of the distribution build jar>
- Running Jumbune Agent
java -jar <jumbune agent jar> <port> <|verbose>
- Running Jumbune
./startWeb
or
./runCli
Code examples are packages inside the distribution,
- For Flow analyzer: BankDefaulters, ClickStreamAnalysis, USRegionPortouts
- For Profilng: MovieRating
- For Data Validation - script
- Quick Start Guide: http://bit.ly/1mY9qWe
- Installation Guide: http://bit.ly/1kBG4Qo
- Release Notes: http://bit.ly/TkWSj8
- Architecture Guide: http://bit.ly/UgJayB
- Deployment Planning Guide: http://bit.ly/1oiXGk2
- Administration Guide: http://bit.ly/1uq5LVV
- Troubleshooting Guide: http://bit.ly/1pXEYOd
- Getting Involved Guide: http://bit.ly/1i7XjDn
Jumbune is licensed under LGPLv3 license