This package contains some code in Julia and Matlab for obtaining a weak convergence curve for the Euler-Maruyama method by using the cluster of the University of Geneva, as discussed in my talk on the numerical analysis seminar on March 9, 2021.

The folder Files_cluster contains the code in Julia, with the bash files for the cluster.
The folder Files_laptop contains the same Julia code, as well as a Matlab file for plotting the convergence curve.
The data is stored in the "results" folders.

For running a code on the cluster:
-Connect via ssh to baobab (using WinSCP on Windows for instance),
-Move the folder Files_cluster in your home directory on the cluster,
-Change the number of processors you want for your job in start.sh, and the parameters of your job in script_job_julia.sh,
-Open a terminal window (with Putty on windows for instance) and type "cd Files_cluster",
-Run your job by typing "sh start.sh",
-You can get some information on your job with "sh info.sh",
-When the job is finished, type "sh merge.sh" to merge the data files in one file data.txt,
-After you copied the data on your personal computer, you can clean all the files with "sh clean.sh",
-You can stop your job at any moment with the command "scancel job_identification_number".

For any further detail, and if you want to obtain the link to the zoom recording of the talk, send me an email at Adrien.Laurent@unige.ch.