jerome
05eccd5b53
- Moved scattered scripts, templates, and documentation into organized directories (applications/, scripts/, assets/). - Updated .gitignore to strictly exclude secrets, state files, and IDE configs. - Added comprehensive README.md outlining repository structure and best practices. - Preserved all existing documentation and technical architecture files. - Prepared infrastructure/ for AWS Org and Proxmox Terraform management.
47 lines
1.5 KiB
Markdown
47 lines
1.5 KiB
Markdown
# 01 - Proxmox & VM Passthrough — Home
|
|
**Purpose:** Complete setup guide for Proxmox VM passthrough with optimal resource allocation.
|
|
|
|
## Topics Covered
|
|
|
|
- **GPU Passthrough:** Configure AMD/Intel/NVIDIA GPUs for VM access
|
|
- **NVMe Passthrough:** Pass NVMe drives to VMs for high I/O performance
|
|
- **USB Passthrough:** Connect USB devices (webcams, dongles, etc.) to VMs
|
|
- **Networking:** Bridge networking, VLANs, and virtual switches
|
|
- **Resource Allocation:** CPU cores, RAM, and storage optimization
|
|
|
|
## Related Guides
|
|
|
|
- **[00 - Proxmox & VM Passthrough](00 - Proxmox & VM Passthrough.md)** — This overview document
|
|
- **[03 - AMD GPU Passthrough](../03\ -\ AMD\ GPU\ Passthrough/)** — AMD-specific setup
|
|
|
|
## Quick Start
|
|
|
|
### Create a VM with GPU Passthrough
|
|
|
|
```bash
|
|
# Step 1: Find available GPU
|
|
rocm-smi | grep "GPU" # AMD
|
|
nvidia-smi # NVIDIA
|
|
|
|
# Step 2: Create VM with GPU passthrough
|
|
qm create 100 --name qwen-vm --memory 32768 --cores 8 --arch x86_64
|
|
qm set 100 --hostpci0=0000:03:00.0,pcie=1,x-vga=1 # AMD example
|
|
|
|
# Step 3: Install QEMU Guest Agent
|
|
apt install qemu-guest-agent
|
|
|
|
# Step 4: Reboot VM
|
|
qm reboot 100
|
|
```
|
|
|
|
## Best Practices
|
|
|
|
1. **IOMMU Groups:** Always check if devices are in separate IOMMU groups
|
|
2. **VFIO Modules:** Ensure `vfio-pci` module is loaded before boot
|
|
3. **PCIe Hotplug:** Disable for stability unless required
|
|
4. **Memory Ballooning:** Disable in VM for consistent performance
|
|
|
|
---
|
|
|
|
*This guide is essential for GPU passthrough to VMs running LLM inference workloads.*
|