| Sr. No. | Syllabus | Duration | 40 Hours | 5 days x 8 Hours |
| 1 | Foundations & Installation | Day 1 | | |
| | HPC concepts and workload management. | | | |
| | SLURM overview: architecture, components (slurmctld, slurmd, slurmdbd). | | | |
| | Comparison with other schedulers (PBS, LSF). | | | |
| | Environment preparation: OS installation, networking, NTP, kernel tuning, security basics. | | | |
| | Installing prerequisites: Munge authentication, development tools, Munge Key | | | |
| | Building/Installing SLURM from repositories. | | | |
| 2 | Core Configuration | Day 2 | | |
| | Understanding slurm.conf structure | | | |
| | Node definitions, partitions, parameters. | | | |
| | Configuring accounting database (slurmdbd). | | | |
| | Initial testing of SLURM setup. | | | |
| | Setting up job submission and queues. | | | |
| 3 | Job Scheduling & Resource Management | Day 3 | | |
| | Job submission scripts and resource requests (CPU, memory, GPU) | | | |
| | SLURM commands: srun, sbatch, squeue, scancel. | | | |
| | Advanced scheduling: reservations, priorities, fair‑share scheduling | | | |
| | Configuring QoS (Quality of Service). | | | |
| | Troubleshooting job failures. | | | |
| | Log Analysis | | | |
| | | | | |
| 4 | Advanced Features & Monitoring | Day 4 | | |
| | SLURM accounting and reporting. | | | |
| | Integrating with MPI/OpenMP workloads. | | | |
| | GPU jobs scheduling | | | |
| | Monitoring tools: sacct, sstat, scontrol. | | | |
| | Performance tuning and scaling strategies. | | | |
| | IOR Benchmarking | | | |
| | Application Benchmarking | | | |
| 5 | Administration & Troubleshooting | Day 5 | | |
| | High availability (HA) setup for SLURM controllers | | | |
| | Backup and disaster recovery strategies. | | | |
| | Common troubleshooting scenarios. | | | |
| | Security hardening and best practices. | | | |
| | SLURM Upgrade | | | |
| | DB Restore Process | | | |
| 6 | Hands‑on labs | | | |
| | Deploying a small SLURM cluster, running workloads. | | | |
| | Submitting jobs to SLURM scheduler | | | |
| | scontrol – job output and logs anslysis | | | |
| | Final assessment and certification. | | | |
| | | | | |
| 8 | Workshop Format | | | |
| | Duration: 40 hours (5 days × 8 hours). | | | |
| | Frequency: 2 sessions per week (Thursdays & Fridays) | | | |
| | OR Weekends – (Saturday & Sunday) | | | |
| | Mode: Online or onsite, theory + hands‑on labs | | | |
| | Audience:HPC admins, lab engineers, technical architects, DevOps professionals. | | | |
| | Ideal for beginners or those new to HPC clusters. | | | |
