Skip to content

Latest commit

 

History

History
74 lines (44 loc) · 1.68 KB

README.md

File metadata and controls

74 lines (44 loc) · 1.68 KB
Raw

Distributed Graph Processing

Algorithms to analyse and process graphs in a distributed way using Apache Hama or Apache Giraph.

Requirements

  • Git
  • Java 7
  • Maven 3

Installing necessary libs

To install both libs Apache Hama Graph(PATCHED) and Apache Giraph 1.1.0:

sh lib/install_libs.sh

If you only want to use one of the libs:

Apache Giraph 1.1.0

sh lib/install_giraph.sh

or,

git clone https://github.com/apache/giraph.git -b release-1.1

cd giraph

mvn install -DskipTests

Apache Hama Graph - Patched

sh lib/install_hama_patched.sh

or,

mvn install:install-file -Dfile=libs/hama-graph-0.6.4-PATCHED.jar -DgroupId=org.apache.hama -DartifactId=hama-graph -Dversion=0.6.4-PATCHED -Dpackaging=jar

Building

After the necessary libs were installed just run:

mvn package

Algorithm Example(Runner)

java -jar <PATH>/algorithm-example.jar alg plat -i <input file> -o <output path> -w <number of workers>

  • plat: Defines if it will run in Giraph or Hama. Possible values: hama, giraph.
  • alg: The algorithm that will run.
  • input file: DFS path to the input file.
  • output path: DFS path to the output file.
  • number of workers: Number of workers in Hama and Number of the
  • l flag: To run the algorithm locally, without needing to configure Hama or Giraph.

java -jar <PATH>/algorithm-example.jar -h

  • h flag: Prints possible options.

Input

The input being used has the following format:

verticeId verticeValue targetVerticeId edgeValue target2VerticeId edge2Value ....

vertice2Id verticeValue targetVerticeId edgeValue target2VerticeId edge2Value ....

...