HPC102: Running a Simple HPC Job

This hands-on webinar introduces participants to the practical workflow for running batch jobs on the UTS Cetus HPC. Participants will learn how to connect to Cetus, use essential Unix commands, and make simple edits to job files using a terminal-based text editor.

The course then focuses on writing PBS job scripts, including how to request CPU cores, memory and walltime, set up working directories, use scratch storage appropriately, and avoid running large computations on the login node. Attendees will work through example scripts and understand how PBS schedules jobs onto compute nodes.

By the end of the session, participants will be able to submit jobs, monitor progress, cancel jobs when needed, and review output files after completion. The webinar is designed for researchers who are new to PBS job submission and want a clear, practical starting point for using Cetus HPC.

Objectives

The training aims to enable attendees to work with confidence on UTS Cetus HPC and to:

• Use key Unix commands for basic file navigation.
• Create and edit simple PBS job scripts.
• Estimate CPU, memory and walltime requirements.
• Submit jobs to the PBS scheduler.
• Monitor job status and interpret common states.
• Cancel queued or running jobs when required.
• Review output and error files after completion.

Prerequisites

The only prerequisite for this course is that you have a UTS Cetus user account ready for login. If you do not have a Cetus account you may register for this course; however, you will not be able to take full advantage of any hands-on exercises.