Tutorials

We prioritize quality over volume. Expect living documents, diagrams, and copy/paste-ready code blocks rather than auto-generated man pages. Each tutorial is tested against current toolchains (Slurm 24.05, Apptainer 1.3, Open MPI 5.x) and the security controls Fortune 500 enterprises demand.

• Designing Resilient Slurm Jobs

  Blueprints for multi-step HPC workflows that survive busy queues, preemption, and node hiccups.

  • Intent-first job headers
  • Heterogeneous allocations
  • Observability hooks

• MPI Launch Patterns That Scale

  Modern MPI playbook covering multi-NIC fabrics, collectives offload, and debugging across thousands of ranks.

  • UCX/libfabric tuning
  • Flux/Balsam orchestration
  • Profiling workflow

• Containers on Regulated HPC Clusters

  Run Apptainer, OCI, and NVIDIA NGC images with zero-trust controls, GPU access, and reproducible pipelines.

  • Build & signing workflow
  • Policy enforcement
  • Troubleshooting checklist