Writing PBS Job Scripts for HPC Systems

High-performance computing clusters offer immense computational power, but accessing these resources requires knowing how to communicate with the system scheduler. This course provides a clear, step-by-step introduction to writing and executing Portable Batch System (PBS) job scripts to run your programs efficiently. You will transition from running simple local commands to orchestrating complex computational tasks on shared academic or corporate clusters. By understanding how to request resources properly, you will avoid common scheduling delays and run your analyses with confidence. What you'll learn: - Understand the core architecture of high-performance computing systems and the role of the PBS scheduler - Configure resource requests including CPU cores, memory limits, and walltime allocations - Write clean, portable shell scripts utilizing PBS directives to automate your workflows - Manage environment variables and directory structures inside a cluster environment - Submit, monitor, and troubleshoot jobs using standard command-line tools like qsub, qstat, and qdel - Apply modern best practices for log management, error handling, and resource efficiency This course begins with foundational concepts of cluster computing and PBS scheduler mechanics. You will then progress through structured text-only explanations and practical script templates, learning to build, submit, and debug your own job configurations. This course is designed for researchers, data scientists, and software developers who are new to high-performance computing. No prior cluster experience is required, though a basic familiarity with the Linux command line is helpful. Start writing efficient PBS scripts and unlock the full power of high-performance computing today.