|
| 1 | +--- |
| 2 | +title: Quick Start |
| 3 | +--- |
| 4 | + |
| 5 | +# Quick Start |
| 6 | + |
| 7 | +This guide walks you through accessing the SMU **M3 Cluster** and running your first job using either **SSH** or the **OnDemand Portal**. |
| 8 | + |
| 9 | +--- |
| 10 | + |
| 11 | +## Accessing the M3 Cluster via SSH |
| 12 | + |
| 13 | +> Remember to replace example terms (like `username` and `slurm_account_name`) with your own credentials. |
| 14 | +
|
| 15 | +### 1. Open a Terminal |
| 16 | + |
| 17 | +- **macOS/Linux:** Open **Terminal** |
| 18 | +- **Windows 11+:** Open **PowerShell** |
| 19 | + |
| 20 | +### 2. Connect to M3 |
| 21 | + |
| 22 | +```bash |
| 23 | +ssh username@m3.smu.edu |
| 24 | +``` |
| 25 | + |
| 26 | +Enter your SMU password (input will not show on screen) |
| 27 | + |
| 28 | +Authenticate with DUO by pressing 1 for your associated device |
| 29 | + |
| 30 | +### 3. Create a Job Script |
| 31 | + |
| 32 | +Create a new job file using nano: |
| 33 | + |
| 34 | +```bash |
| 35 | +nano my_first_job.sbatch |
| 36 | +``` |
| 37 | + |
| 38 | +Paste the following contents into the editor: |
| 39 | + |
| 40 | +```bash |
| 41 | +#!/bin/bash |
| 42 | +#SBATCH -A slurm_account_name |
| 43 | +#SBATCH -J my_first_job |
| 44 | +#SBATCH -o my_first_job_%j.out |
| 45 | +#SBATCH -e my_first_job_%j.err |
| 46 | +#SBATCH -p dev |
| 47 | +#SBATCH -t 0-00:30:00 |
| 48 | +#SBATCH --mem=4G |
| 49 | +#SBATCH -c 1 |
| 50 | + |
| 51 | +# Wait for 100 seconds so you can check the queue using 'squeue --me' |
| 52 | +sleep 100 |
| 53 | + |
| 54 | +# Print a simple message |
| 55 | +echo "Hello, World!" |
| 56 | +``` |
| 57 | + |
| 58 | +Save and exit nano: |
| 59 | + |
| 60 | +``` |
| 61 | +Ctrl + X → Y → Enter |
| 62 | +``` |
| 63 | + |
| 64 | +### 4. Submit and Monitor Your Job |
| 65 | + |
| 66 | +Submit your job: |
| 67 | + |
| 68 | +```bash |
| 69 | +sbatch my_first_job.sbatch |
| 70 | +``` |
| 71 | + |
| 72 | +Check the job status: |
| 73 | + |
| 74 | +``` bash |
| 75 | +squeue --me |
| 76 | +``` |
| 77 | + |
| 78 | +Check the outputs: |
| 79 | + |
| 80 | +``` bash |
| 81 | +cat my_first_job_<jobid>.out |
| 82 | +cat my_first_job_<jobid>.err |
| 83 | +``` |
| 84 | + |
| 85 | +## Accessing the M3 Cluster via OnDemand |
| 86 | + |
| 87 | +Connect to PerunaNet or the SMU campus network. |
| 88 | + |
| 89 | +1. Open a browser and visit: |
| 90 | +https://hpc.smu.edu |
| 91 | + |
| 92 | +2. Log in using your SMU credentials. |
| 93 | + |
| 94 | +3. From the top menu, select My Interactive Sessions. |
| 95 | + |
| 96 | +4. On the left sidebar, choose an interactive app (e.g., RStudio, Jupyter, etc.). |
| 97 | + |
| 98 | +5. Select your SLURM account and job settings (defaults are fine for testing). |
| 99 | + |
| 100 | +6. Click Launch and wait for the job to start in the queue. |
| 101 | + |
| 102 | +7. Once ready, click Launch again to open the session in your browser. |
| 103 | + |
| 104 | +When finished, delete the job from My Interactive Sessions to free resources. |
| 105 | + |
0 commit comments