How to Run Lumerical on Clemson’s Palmetto Cluster
This is a guide to get you started on the Palmetto Cluster. There are a few ways of using it. I’ll show you how I use it here:
The very first step is to make sure you have an account on the Palmetto Cluster. Follow this link to know how to do that: https://www.palmetto.clemson.edu/palmetto/basic/new/
I use the MobaXTerm client to log onto the cluster on my Windows PC.
In MobaXTerm, create a new session. Set server address to “login.palmetto.clemson.edu” and port to 22. Use your username and password to login. Use Duo Prompt (“1”) as necessary.
In the command window, type “whatsfree” to check available resources.
To use an available compute node, type:
qsub -I -X -l select=1:ncpus=48:mem=372gb,walltime=72:00:00
This will choose the 48 CPU, 372 GB memory compute node for 72 hours. Modify the command based on available resources to use the desired compute node.
After compute node is ready, you would see the following output:
After job is ready, you can now run Lumerical on it. To load the module, type:
module load lumerical/2020a
Next, To run MODE, type:
mode-solutions
To run FDTD, type:
fdtd-solutions
You can use the file explorer in MobaXTerm to transfer files between your PC and the cluster.
If you are running on a macbook or windows laptop, you cannot close the lid, as it will stop the simulation. This can be fixed in a few ways.
- You can use the Palmetto Login VM. This is a VM computer where the Lumerical GUI will open. You login to the VM first, then open terminal and run above commands to run the compute node. You can leave the VM and log in later to see it open and running.
- I use this. I run palmetto via MobaXTerm on my home or office PC – which is turned on 24/7. I remote login to the PC with my macbook to check status or make changes. Palmetto is running on my own PC so I have more control here. The login VM sometimes becomes buggy.
