> For the complete documentation index, see [llms.txt](https://docs.vergeos-demo.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.vergeos-demo.com/run-the-platform/system-administration/generic-pci-passthrough.md).

# One-to-One PCI Passthrough

One-to-One PCI passthrough involves giving a single VM at a time, access to a PCI device that is physically attached to a host node. The VM guest operating system utilizes the device as if it were directly attached.

## Host Installation/Configuration

{% hint style="danger" %}
**It is important to read and be familiar with** [**PCI Passthrough Risks and Precautions**](/run-the-platform/system-administration/device-pass-overview.md#pci-passthrough-risksprecautions) **before making passthrough configurations.**
{% endhint %}

1. Navigate to the **Resource Manager Dashboard** (**Infrastructure** > **Resources** from the top menu **-OR-** Main Dashboard > Resources) **-OR-** Navigate to a **specific node dashboard** (**Infrastructure** > **Nodes** from the top menu \*-OR-**Main Dashboard > Nodes**, then double-click desired node in the list.)
2. Click **PCI Devices**.
3. Select the desired device(s) in the list and click **Make Resource** on the left menu.
4. **Select an existing PCI device resource group** from the list **-OR-** **Create a new PCI resource group**.

### Creating a PCI Devices Resource Group

1. **Name**: label used to identify the resource group (i.e. device pool); use a descriptive name so that users are easily able to identify the type of device(s) in this group, e.g. *PCI GPUs*
2. **Type**: ***PCI***
3. **Description**: optional field to provide more administrative text about the resource group.
4. **Class**: select the most appropriate classification from the dropdown list. This field is only used to apply an associated icon to the resource group, and does not affect functionality.
5. Click **Submit** to finish creating the new resource group.
6. **A reboot of the associated node(s) may be necessary**; a message will appear at the top of the Resource Manager dashboard.

{% hint style="warning" %}
**Follow proper** [**Maintenance Mode**](/run-the-platform/operations/maintenance-mode.md) **procedures when rebooting a node to avoid workload disruptions.**
{% endhint %}

After the resource group is selected/created, a *Success* message should appear indicating resource rules were created for the device(s).

## VM/Guest Configuration

1. Navigate to the dashboard of the desired VM (**Virtual Machines** > **List** > double-click desired VM in the listing).
2. Click **Devices** on the left menu.
3. Click **New** on the left menu.

{% hint style="success" %}
**You can also attach devices to a VM via Resource Manager; this method allows for adding multiple devices to the VM at once. From Resource Group dashboard > double-click the desired PCI Group > View Machine Devices > New; select the VM from the Machine dropdown list.**
{% endhint %}

4. Device Entry Form fields:

* **Name**: provide a name to adequately identify the type of device(s), or leave blank to allow the system to auto-generate a device name.
* **Type**: ***PCI***
* **Description**(optional): additional text can be entered here for administrative purposes.
* **Resource Group**: select the appropriate PCI resource group from the dropdown list
* **Count**: indicate the number of PCI devices to attach to the VM (from this resource group).

5. Click **Submit** to complete adding the new device.
6. The VM will need a **restart** in order to attach the device. From the VM dashboard, click the *Restart* link on the message that appears at the top of the dashboard, or click **Restart** on the left menu.
7. Install any required client **drivers**. Consult hardware vendor documentation.
8. **Restart the VM**.

## Pass through a PCI Device to a Tenant

PCI devices can be passed to a tenant for the tenant to pass to its own VMs. When you pass through PCI devices to the tenant, a new resource group is created within the tenant.

{% hint style="info" %}
**When devices are shared to a tenant, they are thick provisioned (i.e. the tenant then owns the devices, so they cannot be assigned to other VMs or tenants even when not in use.)**
{% endhint %}

1. Navigate to the desired **tenant dashboard** (from the top menu: Tenants > List > double-click the tenant within the list.)
2. Click **Nodes** on the left menu.
3. **Double-click one of the tenant nodes**.
4. Click **Devices** on the left menu.
5. Click **New** on the left menu.
6. Enter a **Name** or leave blank to allow the system to auto-create a name; this name will be used for the resource group created in the tenant.
7. **Type**: ***PCI***
8. **Description**: provides a place to store additional administrative information about the device pool.
9. **Count**: number of devices to pass into the tenant resource group.
10. When fields are completed, click **Submit** to finish passing to the tenant.
11. The device(s) will now be available as a resource group to attach to tenant VMs. Follow [**VM/Guest Configuration**](#vmguest-configuration) instructions above. In order to use the passthrough device, the VM must run on the tenant node where the device is attached.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.vergeos-demo.com/run-the-platform/system-administration/generic-pci-passthrough.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.