tae-jun / hadoop-cuda-chart

Hadoop CUDA performance chart - Parallel Software Design Lab, University of Seoul

Home Page:parlab.uos.ac.kr

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Hadoop Job History Chart

Parallel Software Design Lab, University of Seoul

Requirements

Build

git clone https://github.com/tae-jun/hadoop-cuda-chart.git
cd hadoop-cuda-chart
sudo make install
make

Usage

1. Run web server
cd server
node src/app.js

NOTE: Your command may need to be profixed with sudo

2. Start history server
$HADOOP_HOME/sbin/mr-jobhistory-daemon.sh start historyserver

NOTE: You better execute this command on namenode

3. Run client to use history server REST API
cd client
node src/app.js [options]

NOTE: Client must be on same network area with namenode ######Options

  • [-h | --host] web server address
  • [-p | --port] web server port
  • [--hHost] history server address
  • [--hPort] history server port

NOTE: You can also change options on "client/src/config.json" file

4. Run Web application

http://{ WEB_SERVER_HOSTNAME }:10000/

NOTE: You better open with CHROME browser

About

Hadoop CUDA performance chart - Parallel Software Design Lab, University of Seoul

parlab.uos.ac.kr


Languages

Language:TypeScript 71.5%Language:HTML 12.3%Language:JavaScript 11.0%Language:CSS 4.1%Language:Makefile 1.1%