Operation Manual

Setup:

1. Download the files from GitHub by entering the following command:

   git clone https://github.com/jorgeorodriguez/ISTA520_midterm.git
   

2. Download CCTools 4.0.2 & Installing

   wget http://www.nd.edu/~ccl/software/files/cctools-4.0.2-source.tar.gz
   
   cd ~/cctools-4.0.2-source
   
   ./configure --prefix ~/cctools && make install
   

3. Set environment variables by running the following commands:

   export PATH=~/cctools/bin:${PATH}
   
   export PYTHONPATH=~/cctools/lib/python2.4/site-packages:~/cctools/lib/python2.6/site-packages:${PYTHONPATH}
   

Get started:

1. Run the following script to split the files into sections for benchmarking:

   python splitting_densityGrid.py
   

2.  Now we need to take take the IDs from the grav_pos.txt file:
   
    python split_grav_pos.py
   

3. We can now run workqueue to give jobs to workers. This workqueue needs prism.py and grav_per_point.py so make sure they are in the directory

   python grav_wq.py
   

4. Next, we send workers to start processing the tasks. This command will submit a worker to lima with a timeout period of 10 seconds, using port number 54601. The last value will specify how many workers you would like to send for this job

   torque_submit_workers -d all -t 10 lima.futuregrid.org 54601 1
   

5. The grav_wq.py script will generate several output files that will need to be combined into one. For that we can run:

   merge.output.py
   

6. Finally, we can generate a checksum on the final output file to make sure it maintains its integrity:

   md5sum finalfile.txt