Tracks

Instead of dumping hundreds of disconnected tutorials, we map learning tracks to real workflows inside a modern HPC organization. Every link points to a living document—kept current with the latest scheduler releases, chip roadmaps, and compliance rulings.

• Access & Environment Readiness

  Bastions, SSH policies, storage staging, and reproducible environments so researchers can land jobs quickly without compromising security.

  • HPC Foundations for 2025 Clusters
  • SSH posture for HPC admins
  • Container-ready clusters (Apptainer, NGC)

• Scheduler Craft

  Slurm-focused today (24.05 release) with Flux/PBS coverage coming soon. These guides keep utilization high without torching QoS.

  • Reliability Playbook for Busy Slurm Queues
  • Designing resilient Slurm jobs
  • Automated readiness checklists

• GPU & AI Convergence

  Making accelerators productive—from MIG slicing and NVLink monitoring to container orchestration and compliance around sensitive datasets.

  • Advanced Techniques for GPU-Heavy Science
  • MPI launch patterns that scale
  • Containers on regulated clusters

• Operations & Observability

  Node health scoring, telemetry, patch windows, and communication rituals that keep regulators, researchers, and finance aligned.

  • Operational KPIs checklist
  • Search + knowledge base strategy
  • Contact the field team