Overview

The AI and Simulation Lab (ASL) Server provides a platform for high-performance computing tasks so researchers and students can run computationally intensive jobs on advanced GPU and CPU resources.

Server access

Use the original SSH command from the mirrored page:

ssh -p 44065 [Your Username]@203.159.10.44

Remark from the original page: if you want to access ASL Slurm from off-campus, connect to AIT-VPN first.

System specifications

• ASL-gpu (Default Partition): 4 x NVIDIA A6000 GPUs for deep learning and GPU-accelerated jobs.
• ASL-cpu: 128 V-CPUs for CPU-based processing and parallel workloads.

Getting started with Slurm

1. Create a Slurm batch file using the guide now available on the Intro-SLURM-Script package page.
2. Watch the original tutorial reference: YouTube tutorial by Dr. Chaklam.

Video tutorial

Additional tips

• Use commands such as sbatch, squeue, and scancel to manage jobs.
• Set the proper partition in your script, for example #SBATCH --partition=ASL-gpu.