Excercise for ECOL-346 Bioinformatics class
https://ood.hpc.arizona.edu/pun/sys/dashboard
- Choose 1 hour, 1 core, standard queue
- Make sure your account is correct
- Connect to Jupyter
We are doing this so we don't get slowed down on the login node.
Use a Bash kernel Jupyter Notebook if you want to have notes to look back on!
git clone https://github.com/agladstein/ECOL-346-HPC-demo.git
cd ECOL-346-HPC-demo
git clone https://github.com/agladstein/SimPrily.git
python SimPrily/simprily.py --help
Load the Singularity module
module load singularity
Use Singularity to pull the Image
cd SimPrily
singularity pull docker://agladstein/simprily
Try SimPrily with the Singularity container
singularity exec simprily.simg python /app/simprily.py --help
ALWAYS TEST SMALL FIRST!_
cd SimPrily
python simprily.py -p examples/eg1/param_file_eg1.txt -m examples/eg1/model_file_eg1.csv -i test -o output
If you had to use the Singularity containter, run like this:
singularity exec simprily.simg python /app/simprily.py -p examples/eg1/param_file_eg1.txt -m examples/eg1/model_file_eg1.csv -i test -o output
Make sure you have results in:
output/results/
Open simprily_simple.pbs, and change group
to your group.
What queue are we running on?
How many cores are we asking for?
How many wall hours are we asking for?
cd
back to ECOL-346-HPC-demo
and submit the job
qsub simprily_simple.pbs
Check the status of the queue
qstat -u [user-name]
Once it completed, check your simprily1.o*
file for any errors, and check SimPrily/output/results/
You might need to use the singularity container. If you do, change the pbs script as needed.
Open simprily_batch.pbs, and change group
to your group.
What is different about this pbs script?
How are we running 1000?
How many cores are we asking fo?
How many wall hours are we asking for?
Submit the job
qsub simprily_batch.pbs
Check the status of the queue
qstat -u [user-name]
Check the progress of your batch job
qstat -t [jobid]
qstat -f [jobid]
Once it completed, check one of the simprilyB.o*
file for any errors, and check output/
for results
Open simprily_parallel.pbs, and change group
to your group.
What is different about this pbs script?
How are we running 1000?
How many cores are we asking for?
How many wall hours are we asking for?
Submit the job
qsub simprily_parallel.pbs
Check the status of the queue
qstat -u [user-name]
Check the progress of your batch job.
How many jobs do you expect?
qstat -t [jobid]
qstat -f [jobid]
Once it completed, check one of the simprilyP.o*
file for any errors, and check output/
for results
For this use case, which is better - single core or multi core?
Combine the results into one file.