PegasusN runs on any machine that supports hadoop and spark. However, the shell scripts and code packaging scripts works easily in Linux or Unix machines. PegasusN needs the following softwares to be installed in the system:
- Hadoop version 2.x or greater
- Spark version 2.1.x or greater
- Java 1.8.x or greater, preferably from Sun
Download and Install
Installation is as simple as downloading and extracting the installation file.
- Download the file pegasusn-3.0.tar.gz.
- Extract the file by running the following command: tar xvzf pegasusn-3.0.tar.gz. Then, the directory 'pegasusN' will be created.
- cd pegasusN
- To build the package, you may type `tools/sbt assembly` command.
PegasusN can be run in one of the two ways: by using an interactive shell `pegasus_hadoop.sh` or `pegasus_spark.sh`, or by running shell scripts for each algorithm. For first-time users, we recommend to use the interactive shell. Advanced users may want to use shell scripts for each algorithm.
PegasusN Interactive Shell
- Type `bin/pegasusn_hadoop.sh` or `bin/pegasusn_spark.sh` in the installation directory. The shell for PegasusN appears.
- Type `help` command for available commands in the shell.
- Type `demo` to see the demo. The demo first adds a graph file 'catepillar_star.edge' by the name 'catstar' in PegasusN, computes the inout degree, and generates the degree distribution plot 'catstar_deg_inout.eps'.
Running Each Algorithm
In the installation directory, type 'bin/demo.sh' to see the demo of computing radii and diameter of an example graph using the shell script for running radius algorithm.
The User Guide provides information on the general overview, installation, and supported functions of PegasusN.