Introduction to PBS Job Scheduling for HPC Systems

High-performance computing (HPC) clusters offer immense computational power, but accessing these resources requires understanding how to communicate with a workload manager. The Portable Batch System (PBS) is one of the most widely used job schedulers in scientific and enterprise computing. This course provides a clear, structured introduction to navigating HPC environments and submitting jobs efficiently without needing prior cluster experience. You will transition from a beginner to a confident HPC user who understands how to request CPU, memory, and runtime resources effectively. By learning the mechanics of job submission, you will avoid common scheduling pitfalls and ensure your computational tasks run smoothly and efficiently. What you'll learn: - Understand foundational HPC architecture and the essential role of a job scheduler - Write robust PBS job scripts using standard directives and environment variables - Submit, monitor, modify, and delete jobs using core PBS command-line utilities - Request specific hardware resources including multi-core nodes, memory, and modern GPU accelerators - Configure basic job arrays to automate repetitive, large-scale computational tasks - Troubleshoot failed jobs by analyzing standard output, error logs, and exit codes This text-based course begins with essential HPC terminology, explaining how login nodes, compute nodes, and schedulers interact. You will then progress through step-by-step written guides that demonstrate how to draft job scripts, manage queues, and optimize resource requests for both serial and parallel workloads. This course is designed specifically for researchers, data scientists, engineers, and students who are new to high-performance computing and need to run software on a PBS-managed cluster. No prior HPC or scheduling experience is required, though a basic familiarity with the Linux command line is helpful. Start reading today to unlock the full computational power of your HPC cluster.