> For the complete documentation index, see [llms.txt](https://docs.stg.ai.neevcloud.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.stg.ai.neevcloud.com/ai-storage/accessing-storage.md).

# Accessing Storage

NeevCloud provides multiple ways to access your storage, depending on whether you prefer command-line interfaces or browser-based interactive environments.

## Default Mount Path

All storage volumes automatically mount at `/data` on your GPU instance. This standardized path simplifies your workflows:

```bash
# Your storage is always available at:
cd /data

# List your data:
ls -lh /data

# Check storage usage:
df -h /data
```

You don't need to configure mount points or fstab entries—everything is ready to use when your instance starts.

## SSH Access

SSH gives you direct terminal access to your GPU instance and its attached storage. This is ideal for running training scripts, managing files, or executing command-line tools.

### Connection Syntax

```bash
ssh -i /path/to/key-pair-name.pem -p <PORT> root@<IP_ADDRESS>
```

### Example

```bash
ssh -i ~/.ssh/neevcloud-key.pem -p 30890 root@192.168.10.20
```

**What each parameter means:**

* `-i /path/to/key-pair-name.pem`: Path to your private SSH key file
* `-p 30890`: SSH port number (you'll get this from your instance details)
* `root@192.168.10.20`: Username and IP address of your instance

Once connected via SSH, you can:

```bash
# Navigate to your storage
cd /data

# Upload datasets using scp from your local machine
scp -i ~/.ssh/neevcloud-key.pem -P 30890 dataset.tar.gz root@192.168.10.20:/data/

# Download model checkpoints to your local machine
scp -i ~/.ssh/neevcloud-key.pem -P 30890 root@192.168.10.20:/data/model_checkpoint.pth ./

# Run your training script
python /data/train.py --config /data/config.yaml

# Check storage usage
df -h /data
du -sh /data/*
```

## Jupyter Notebook Access

Jupyter Notebook provides an interactive browser-based environment perfect for exploratory data analysis, model prototyping, and visualization.

### How to Access

1. Navigate to your instance details from the NeevCloud dashboard.
2. Click the "Open Jupyter Notebook" button.
3. Your browser opens a Jupyter environment with `/data` already accessible.

### Working with Your Storage in Jupyter

```python
import os
import pandas as pd

# Your storage is mounted at /data
data_path = '/data'

# List files in your storage
print(os.listdir(data_path))

# Load a dataset from your storage
df = pd.read_csv('/data/training_data.csv')

# Save results back to your storage
model.save('/data/checkpoints/model_epoch_10.h5')

# Check available space
import shutil
total, used, free = shutil.disk_usage('/data')
print(f"Free space: {free // (2**30)} GB")
```

**Use Jupyter Notebook when you:**

* Need to visualize data or training metrics.
* Want to interactively explore datasets.
* Need to document your workflow with markdown and code.

## Comfy UI Access

Comfy UI is a browser-based interface specifically designed for AI workflows, particularly image generation and processing pipelines.

### How to Access

1. Go to your instance details from the NeevCloud dashboard.
2. Click the "Open Comfy UI" button.
3. Your browser launches the Comfy UI environment.

### Working with Storage in Comfy UI

* Access your models, datasets, and outputs directly through the UI.
* Your `/data` directory is available for loading custom models.
* Save generated outputs to `/data` for persistence.

**Use Comfy UI when you:**

* Are working with diffusion models or image generation.
* Need a visual workflow builder.
* Want to quickly iterate on AI pipelines without writing code.

## Best Practices for Access

### Security

* Keep your SSH private keys secure and never share them.
* Use strong key passphrases.
* Rotate keys periodically for production workloads.

### File Management

* Organize data in `/data` with clear directory structures.
* Use descriptive filenames and directory names.
* Clean up temporary files to avoid filling storage.