MIG Configuration on DGX Station

Configure MIG (Multi-Instance GPU) partitions on the DGX Station GB300.

Steps

Find the GB300 GPU index. Run:

nvidia-smi --query-gpu=index,name --format=csv,noheader

Check current MIG state:

nvidia-smi -i <GB300_INDEX> -q | grep -i "MIG Mode"

If MIG is already enabled, show current instances:
```
nvidia-smi mig -lgi -i <GB300_INDEX>
nvidia-smi mig -lci -i <GB300_INDEX>
```
If the user wants to reconfigure, destroy existing instances first (step 6).

If MIG is not enabled, enable it. All GPU processes must be stopped first:

# Check for running GPU processes
sudo fuser -v /dev/nvidia*

# Enable MIG
sudo nvidia-smi -i <GB300_INDEX> -mig 1

# Verify
nvidia-smi -i <GB300_INDEX> -q | grep -i "MIG Mode"

Show available profiles and help the user choose a layout:

nvidia-smi mig -lgip -i <GB300_INDEX>

Common GB300 MIG profiles:

Profile	ID	Memory	Use case
1g.35gb	19	~35 GB	Small models (7-8B), dev/test
1g.35gb+me	20	~35 GB	Same + media extensions
1g.70gb	15	~70 GB	Slightly larger inference
2g.70gb	14	~70 GB	Medium models (14-30B)
3g.139gb	9	~139 GB	Large models (70B quantized)
4g.139gb	5	~139 GB	Large models, more compute
7g.278gb	0	~278 GB	Full GPU as single instance

Suggest layouts based on the user's workload. Examples:

Two models (70B + 8B): 3g.139gb + 2g.70gb + 2g.70gb → IDs 9,14,14
Many small models: 7 × 1g.35gb → IDs 19,19,19,19,19,19,19
One large model with isolation: 7g.278gb → ID 0

Ask the user what models they want to run before suggesting a layout.

Create (or recreate) instances:

If reconfiguring, destroy existing instances first:

sudo nvidia-smi mig -dci -i <GB300_INDEX>
sudo nvidia-smi mig -dgi -i <GB300_INDEX>

Then create the new layout:

sudo nvidia-smi mig -cgi <PROFILE_IDS> -C -i <GB300_INDEX>

Get the MIG device UUIDs:
```
nvidia-smi -L
```
Note the MIG-<uuid> entries — these are used to target specific MIG instances.

Show the user how to use MIG devices:

# Bare metal
export CUDA_VISIBLE_DEVICES=MIG-<uuid>

# Docker
docker run --gpus '"device=MIG-<uuid>"' ...

Report the final layout to the user with UUIDs and suggested docker commands for each instance.

Disabling MIG

If the user wants to return to full-GPU mode:

# Stop all workloads using MIG instances first
sudo nvidia-smi mig -dci -i <GB300_INDEX>
sudo nvidia-smi mig -dgi -i <GB300_INDEX>
sudo nvidia-smi -i <GB300_INDEX> -mig 0

# Ensure Fabric Manager is running for NVLink re-initialization
sudo systemctl start nvidia-fabricmanager

2.9 KiB Raw Blame History Unescape Escape

MIG Configuration on DGX Station

Steps

Disabling MIG

2.9 KiB

Raw Blame History