🖥️ SLURM Guide

A simple guide to using SLURM (Simple Linux Utility for Resource Management) on KFUPM clusters.

---

Current Cluster Setup

📊 Partition Details

Partition	Purpose	Time Limit	Nodes	GPUs
Normal*	Large Models	24 hours	server02	6x A100
RTX3090	GPU computing	24 hours	jrcai01-02	2x (2xRTX 3090)
LoginNode	Access only	-	jrcai23	Login access

* Default partition

👥 Group Management

• Advisor Groups: Each advisor has a group with their students
• Shared Storage: Groups are hard-limited to 1 TB of shared disk space.
• Job Limits:
  • Normal Partition (default): 1 job limit per group
  • RTX3090 Partition: Groups can submit 1 additional GPU job by specifying this partition

📖 Documentation

🔗 How to Connect

Learn how to connect to the SLURM cluster using:

• SSH Terminal
• Visual Studio Code

⚡ How to Use SLURM

Complete guide covering:

• Monitoring commands
• Job submission
• Data transfer
• Account management

---

📋 Basic SLURM Workflow

graph TD
    A[Write Job Script] --> B[Submit with sbatch]
    B --> C[Job Queued - Status: PD]
    C --> D[Resources Available?]
    D -->|No| C
    D -->|Yes| E[Job Running - Status: R]
    E --> F[Job Complete]
    F --> G[Check Results]

Loading

---

Getting Help

Contact Information:

• System Administrator: Contact JRCAI support team
• Technical Issues: mohammed.sinan@kfupm.edu.sa
• Account Problems: Submit ticket through proper channels

Last Updated: 16/9/2025
By: Mohammed AlSinan (mohammed.sinan@kfupm.edu.sa)

Login Node: (check your email/registration details)