Tutorials

These exercises are to demonstrate common workflows using OSN. You may use your Mac or Linux machine to complete these exerciese or use the binder-based Linux desktop linked below.

The above binder is based on Ubuntu 18.04 LTS and XFCE.

Exercise 1

To use your own machine for this exercise, you will need git, wget, and gnuplot.

Pull down the workflow from GitHub:

$ git clone https://github.com/mghpcsim/osn-tutorial.git

Note

If you are using the above binder, the Git repo has already been cloned

Change directories to the exercise directory:

$ cd osn-tutorial/exercise1

Workflow step 0, download data from OSN:

#!/bin/bash

CMD="wget https://mghp.osn.xsede.org/cis220170-bucket01/anscombe.txt"
echo ""
echo running $CMD
echo ""

$CMD >& /dev/null

ret=$?
if [ $ret -ne 0 ]; then
        echo "wget failed ... HALP "
        echo ""
else
        echo "Data retrieved successfully"
        echo ""
fi

Execute the download data script:

$ ./00-download-data.sh

running wget https://mghp.osn.xsede.org/cis220170-bucket01/anscombe.txt

Data retrieved successfully

Workflow step 1, process the data:

#!/bin/bash

echo "running gnuplot using the anscombe.gnu script"
echo " $ gnuplot < anscombe.gnu"

gnuplot < anscombe.gnu
echo ""
ret=$?
if [ $ret -ne 0 ]; then
        echo "Data processing failed ... HALP "
        echo ""
else
        echo "Data processed successfully"
        echo ""
fi

Run the process data script:

$ ./01-process-data.sh
running gnuplot using the anscombe.gnu script
$ gnuplot < anscombe.gnu

Data processed successfully

Workflow step 2, visualize the results:

#!/bin/bash

unameOut="$(uname -s)"
case "${unameOut}" in
Linux*)     machine=Linux;;
Darwin*)    machine=Mac;;
CYGWIN*)    machine=Cygwin;;
MINGW*)     machine=MinGw;;
*)          machine="UNKNOWN:${unameOut}"
esac
#echo ${machine}

if [ $machine == Mac ]; then
  open *.png
elif [ $machine == Linux ]; then
  ristretto anscombe1.png
else
  echo "I dont know you"
fi

Finally, visualize the data:

$ ./02-viz-data.sh