Jumbune is an open-source project to optimize both Yarn (v2) and older (v1) Hadoop based solutions. It's a Linux based framework to assist in fault free analytic solution development, data quality validation and efficient cluster utilization. Jumbune can decoupled from the Hadoop cluster & run on a separate machine. Being a generic implementatio…
JavaScript Java HTML CSS Shell PHP
Permalink
Failed to load latest commit information.
bin Updated scheduler script May 5, 2015
common the build version has been updated for the SNAPSHOT Jun 23, 2016
datavalidation the build version has been updated for the SNAPSHOT Jun 23, 2016
debugger the build version has been updated for the SNAPSHOT Jun 23, 2016
deploy the build version has been updated for the SNAPSHOT Jun 23, 2016
distribution the build version has been updated for the SNAPSHOT Jun 23, 2016
docker-conf Updated the files with new corrected version Jun 22, 2016
examples the build version has been updated for the SNAPSHOT Jun 23, 2016
execution the build version has been updated for the SNAPSHOT Jun 23, 2016
profiling the build version has been updated for the SNAPSHOT Jun 23, 2016
remoting the build version has been updated for the SNAPSHOT Jun 23, 2016
rumen the build version has been updated for the SNAPSHOT Jun 23, 2016
utilities the build version has been updated for the SNAPSHOT Jun 23, 2016
web the build version has been updated for the SNAPSHOT Jun 23, 2016
.dockerignore JUM-100 Ability to build Jumbune docker image from the latest SNAPSHO… May 18, 2015
.travis.yml Added environment variable May 4, 2015
Dockerfile Updated the files with new corrected version Jun 22, 2016
LICENSE.txt Committed Initial version of Jumbune (1.0.0) Jun 5, 2014
NOTICE.txt Removed dependency of SnakeYaml and updated NOTICE.txt Jul 31, 2015
README.md Updated README.md to expose out fewer ports in Docker May 18, 2015
assembly.xml Fixed bug JUM-77 :- Performance degradation of debugger jobs after in… Feb 13, 2015
pom.xml the build version has been updated for the SNAPSHOT Jun 23, 2016

README.md

jumbune Build Status

Synopsis

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.

Website

http://jumbune.org

Issue Tracker

http://jumbune.org/jira/browse/JUM

Building

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

Installation

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>

  • Deploying Jumbune (in verbose mode)

java -jar <location of the distribution build jar> -verbose

  • Running Jumbune Agent

java -jar <jumbune agent jar> <port> <|verbose>

  • Running Jumbune

./startWeb

or

./runCli

Docker Image (Jumbune + Apache YARN )

  • Building from root of the checked out repository $ sudo docker build –t = "jumbune/pseudo-distributed:1.5.0" .

  • Building from the github.com repository $ sudo docker build –t = "jumbune/pseudo-distributed:1.5.0" github.com/impetus-opensource/jumbune/

  • Getting the automated build from docker registry https://registry.hub.docker.com/u/jumbune/jumbune/

  • Running the built image $ docker run -d --name="jumbune" -h "jumbune-docker" -p 8080:8080 -p 5555 jumbune/pseudo-distributed:1.5.0

Code Examples

Code examples are packages inside the distribution,

  • For Flow analyzer: BankDefaulters, ClickStreamAnalysis, USRegionPortouts
  • For Profilng: MovieRating
  • For Data Validation - script

Documentation

License

Jumbune is licensed under LGPLv3 license