ARC-TS

Job script generator for Slurm

This site assists in creating job scripts for the Slurm workload manager.
This tool is currently experimental at U-M and does not yet perform input validation or bounds checking, although future versions will.
Please send input or feedback to hpc-support@umich.edu
For more details about migrating to Slurm please see our Torque to Slurm guide.
Enter the details of your job in the form below and then click on the make script button below.
After the script is made, you can copy and paste to another file or use the download button to save a local copy as a plain text file.

Select cluster using button or follow documentation link to learn more greatlakes (documentation)
lighthouse (documentation)
Job Name
Account
Job Type
Single Core

Suitable for single CPU core jobs only (not parallel).

Single Node (Multithreaded or SMP)

Suitable for shared-memory multithreaded jobs. Note: these jobs cannot use more than one node and cannot use more cores than the number of cores on a node.

Multiple Node (MPI)

Suitable for distributed parallel jobs that use MPI. May use more than one cluster node.

Memory in Gigabytes
Email

Send yourself an email when the job:

 starts running
 ends successfully
 ends with an error
Use this email address:
Walltime
Days
Hours
Minutes
Directory

Run the job from:

The same directory where it is launched.
Your home directory.
The directory with this name:
Modules

Enter the modules that you would like to load:

Command
The job script

Click on the button to make your script. If you change any of the values in the form above you can re-generate the script by clicking the button again.