Skip to content

Latest commit

 

History

History
43 lines (24 loc) · 1.77 KB

RunningOnDiRAC.md

File metadata and controls

43 lines (24 loc) · 1.77 KB

Running On DiRAC

Follow these instructions to get an account (for the SCRC Project): https://github.com/ScottishCovidResponse/SCRCIssueTracking/wiki/DiRAC-Access

You should receive an email describing how to connect to one of the CSD3 login nodes, with something like (replacing your-user-name with your username):

ssh -l your-user-name login.hpc.cam.ac.uk

Load Java:

module load jdk-8u141-b15-intel-17.0.4-3g6rhr2

John has downloaded gradle from https://gradle.org/release-nightly/ and unzipped it to ~/rds/rds-dirac-dc003/dc-nonw1/gradle. To use it (replacing [your-user-name]):

export PATH=$PATH:/home/[your-user-name]/rds/rds-dirac-dc003/dc-nonw1/gradle/bin

Get the source code from GitHub, and build it with:

git clone https://github.com/ScottishCovidResponse/Covid_Simulation_Model.git
cd Covid_Simulation_Model
gradle install

Important: You must re-run gradle install whenever you update the code. (You can use git pull instead of clone if you have already cloned the repository.)

Set "outputDirectory": "" in your model parameters file, so that extra output directories are not created and the script to combine outputs works.

Then have a look at the "TODO"s in the example slurm_submit file, which can be run with the command:

sbatch slurm_submit

You can check the status of your job using:

squeue -n covid-sim-model

And combine the output from multiple jobs into a single csv file using (replacing jobID, lastArrayIndex and outputFileName with the relevant values):

combineArrayJobOutputs.sh jobID lastArrayIndex outputFileName

(After leaving your ssh session,) copy files back to your local machine by running scp from your local machine, e.g.:

scp [your-user-name]@login.hpc.cam.ac.uk:/home/[your-user-name]/out.csv .