---
title: "NFS"
slug: "nfs"
updated: 2026-09-15T14:13:49Z
published: 2026-09-15T14:13:49Z
canonical: "docs.dataloop.ai/nfs"
---

> ## Documentation Index
> Fetch the complete documentation index at: https://docs.dataloop.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# NFS

## **What is NFS?**

NFS is a network-based file sharing protocol that allows multiple machines to access files stored on a central server as if those files were located on their own local disk.

Think of it as a **shared network drive**.

**Example**

Suppose you have a storage server:

```plaintext
Server IP: 10.172.164.34
Shared Folder: /exports/images
```

---

## How NFS Works in DDOE

When DDOE is configured with an **NFS (Network File System) Storage Driver**, the dataset does not store copies of the files inside DDOE. Instead, it accesses the files directly from the NFS server through the compute cluster.

***NFS Server → Compute Cluster → Storage Driver → Dataset***

1. **An NFS server stores the files**: The NFS server is the actual location where your data resides.
2. **The Compute Cluster mounts the NFS share**: The DDOE Compute Cluster establishes access to the NFS server, making the storage available to workloads running within the cluster.
3. **The Storage Driver exposes the NFS storage to DDOE**: The Storage Driver serves as the connection layer between DDOE and the NFS storage, enabling the platform to access files stored on the server.
4. **Datasets access data directly from the NFS storage:** Datasets created using the NFS Storage Driver can access files directly from the connected NFS storage without requiring the data to be uploaded into DDOE.

---

## Example Configuration

```json
{
  "name": "nfs-storage",
  "type": "nfs",
  "config": {
    "server": "10.172.164.34",
    "path": "/exports/images"
  }
}
```

---

## Advantages and Limitations

| **Advantages** | Limitations |
| --- | --- |
| - **Shared Storage**: All pods and nodes can see the same files. - **Centralized Management**: Data is stored in one location. - **Easy Data Access**: No need to copy data into DDOE. - **Suitable for Large Files** Commonly used for: - **Images** - **Videos** - **Documents** - **Medical datasets** | - **Network Dependent**: If the NFS server is unavailable, data becomes unavailable. - **Performance Limitations**: Can become slower when many users access the same storage simultaneously. |

---

## Create an On-Premises NFS Integration

> [!WARNING]
> **Prerequisites**
> 
> - **Roles and Permissions**: Only users with the Organization **Admin** or **Owner** roles are authorized to create integrations. [Learn more](https://docs.dataloop.ai/docs/org-members-groups#org-role-permissions)
> - **NFS Server Address**: The IP address or the hostname of your NFS server.
> - **Export Path**: The directory path on the NFS server that is exported or shared.

1. Log in to the DDOE platform.
2. From the left-side panel, select **Integrations**.
3. Click **Create Integration** -> **Create Storage Integration**. A pop-up window is displayed on the right-side.
4. **Integration Name**: Provide a name for your NFS Integration.
5. **Provider**: Select **On-Prem** from the list.
6. **Integration Type**: Select **NFS** from the list.
7. **NFS Server Address**: Enter the IP address or the hostname of your NFS server.
8. **Export Path**: Enter the directory path on the NFS server that is exported or shared.
9. Click **Create Integration**. The NFS type integration will be created and listed in the Integration page.