Instance Infrastructure
Hardware Specifications
- CPU: E series CPU instances (Azure)
- GPU Options:
- A100 80GB (default for both modes)
- T4 16GB (available via CLI)
- A100 40GB (available via CLI)
- Memory: 24 vCPUs and 220 GiB RAM per GPU (Production mode, Prototyping is fully customizeable)
- Location: United States (region varies)
Pre-installed Software
- CUDA: Version 12.9
- CUDNN: Version 9.0
- PyTorch: Version 2.7.1
- JupyterLab: Pre-installed
- Docker: Available (see “Docker on Thunder Compute” guide)
- Additional scientific Python libraries (NumPy, Pandas, etc.)
Do not attempt to reinstall CUDA. If compatibility issues arise, upgrade your other dependencies (e.g., PyTorch) rather than downgrading CUDA.
Networking
- Egress/Ingress: 7 Gbps
- IP Address: Dynamic
Port Access
- CLI: Use
tnr connect <instance_id> -t <port>to tunnel ports - VS Code: Use the built-in port forwarding feature
- External access: Use Cloudflared:
cloudflared tunnel --url http://localhost:<port>
Storage
Ephemeral Disk
- All instances use ephemeral storage
- Deleting an instance permanently removes all data on its disk
- Back up important files before deleting an instance. See Using Ephemeral Storage for recommended workflows.
Data Retention
- Ephemeral by default: Instance storage exists only while the instance is active
- Backups required: Copy critical data off the instance before deleting it
- Permanent: Deleting an instance cannot be undone