CAR Mobility Systems Lab Setup Guide¶
Welcome to the OSU CAR Mobility Systems Lab documentation! This guide helps lab members set up their development environment and work effectively on the Ohio Supercomputer Center (OSC).
Lab Compute Dashboard
Track our OSC spending, job health, and per-user breakdowns on the live dashboard. Data refreshes automatically every morning. See how it's built.
Quick Start¶
New to the lab? Follow these steps to get up and running:
| Step | Guide | Description |
|---|---|---|
| 1 | Install WSL2 | Windows users: set up WSL2 (your Linux development environment) |
| 2 | Set up VS Code | Install and configure Visual Studio Code with essential extensions |
| 3 | Get OSC Access | Request your account on the Ohio Supercomputer Center |
| 4 | Connect via SSH | Set up SSH keys and connect to OSC from your machine |
| 5 | Start Developing | Use VS Code's Remote-SSH for seamless development on the cluster |
Why These Tools?¶
Each tool in our stack addresses a specific challenge in the ML research workflow — from writing code locally to running large-scale experiments on OSC's supercomputers.
| VS Code | Git / GitHub | SSH | SLURM | uv / modules | Ray | PyTorch | Copilot / Claude | |
|---|---|---|---|---|---|---|---|---|
| Edit & debug code | Primary IDE and debugger | AI-powered suggestions | ||||||
| Collaborate on code | Version control & pull requests | Shared environment specs | ||||||
| Access HPC clusters | Remote-SSH extension | Secure cluster login | ||||||
| Manage Python envs | Reproducible environments | |||||||
| Run compute jobs | Schedule CPUs & GPUs | |||||||
| Automate pipelines | Execute each step | Python-native GPU pipeline orchestration | ||||||
| Train ML models | Monitor & debug | GPU allocation | Install ML stack | Orchestrate experiments | Neural network framework | Help write & fix code |
Documentation Sections¶
-
Getting Started
WSL2, VS Code, Python environments, and AI coding assistants for local development.
-
OSC Basics
Cluster specs, account setup, SSH, remote development, file transfer, and OnDemand.
-
Working on OSC
SLURM job submission, environment management, pipeline orchestration, and simulators.
-
ML Workflows
PyTorch & GPU setup, PyG, project templates, notebook-to-script, and experiment tracking.
-
Contributing
How this site works, adding pages, and setting up your own documentation site.
-
Assignments
Hands-on tasks for new undergraduates: personal websites, EDA, and OSC onboarding.
-
Resources
Troubleshooting guides, useful links, and OSC support contacts.
Tips¶
Use the search
Press S or / to search the documentation. The search indexes all pages and code blocks.
Dark mode
Click the icon in the header to toggle dark mode.
Edit on GitHub
Found an error? Click the icon on any page to suggest an edit.
Quick Links¶
| Resource | Link |
|---|---|
| Lab Compute Dashboard | osc-usage-dashboard |
| OSC Documentation | osc.edu/resources |
| OSC OnDemand Portal | ondemand.osc.edu |
| OSC Status Page | osc.edu/status |
| Lab GitHub | github.com/OSU-CAR-MSL |