proposalbot
85e1a6ed92
Reviewed-by: Eotvos, Oliver <oliver.eotvos@t-systems.com> Co-authored-by: proposalbot <proposalbot@otc-service.com> Co-committed-by: proposalbot <proposalbot@otc-service.com>
99 lines
4.1 KiB
ReStructuredText
99 lines
4.1 KiB
ReStructuredText
:original_name: cce_10_0141.html
|
|
|
|
.. _cce_10_0141:
|
|
|
|
gpu-beta
|
|
========
|
|
|
|
Introduction
|
|
------------
|
|
|
|
gpu-beta is a device management add-on that supports GPUs in containers. If GPU nodes are used in the cluster, the gpu-beta add-on must be installed.
|
|
|
|
Notes and Constraints
|
|
---------------------
|
|
|
|
- The driver to be downloaded must be a **.run** file.
|
|
- Only NVIDIA Tesla drivers are supported, not GRID drivers.
|
|
- When installing or reinstalling the add-on, ensure that the driver download address is correct and accessible. CCE does not verify the address validity.
|
|
- The gpu-beta add-on only enables you to download the driver and execute the installation script. The add-on status only indicates that how the add-on is running, not whether the driver is successfully installed.
|
|
|
|
Installing the Add-on
|
|
---------------------
|
|
|
|
#. Log in to the CCE console and access the cluster console. Choose **Add-ons** in the navigation pane, locate **gpu-beta** on the right, and click **Install**.
|
|
#. Configure the driver link.
|
|
|
|
.. important::
|
|
|
|
- If the download link is a public network address, for example, **https://us.download.nvidia.com/tesla/470.103.01/NVIDIA-Linux-x86_64-470.103.01.run**, bind an EIP to each GPU node. For details about how to obtain the driver link, see :ref:`Obtaining the Driver Link from Public Network <cce_10_0141__section95451728192112>`.
|
|
- If the download link is an OBS URL, you do not need to bind an EIP to GPU nodes.
|
|
- Ensure that the NVIDIA driver version matches the GPU node.
|
|
- After the driver version is changed, restart the node for the change to take effect.
|
|
|
|
#. Click **Install**.
|
|
|
|
Verifying the Add-on
|
|
--------------------
|
|
|
|
After the add-on is installed, run the **nvidia-smi** command on the GPU node and the container that schedules GPU resources to verify the availability of the GPU device and driver.
|
|
|
|
GPU node:
|
|
|
|
.. code-block::
|
|
|
|
cd /opt/cloud/cce/nvidia/bin && ./nvidia-smi
|
|
|
|
Container:
|
|
|
|
.. code-block::
|
|
|
|
cd /usr/local/nvidia/bin && ./nvidia-smi
|
|
|
|
If GPU information is returned, the device is available and the add-on is successfully installed.
|
|
|
|
|image1|
|
|
|
|
.. _cce_10_0141__section95451728192112:
|
|
|
|
Obtaining the Driver Link from Public Network
|
|
---------------------------------------------
|
|
|
|
#. Log in to the CCE console.
|
|
#. Click **Create Node** and select the GPU node to be created in the **Specifications** area. The GPU card model of the node is displayed in the lower part of the page.
|
|
|
|
3. Visit https://www.nvidia.com/Download/Find.aspx?lang=en.
|
|
|
|
4. Select the driver information on the **NVIDIA Driver Downloads** page, as shown in :ref:`Figure 1 <cce_10_0141__fig11696366517>`. **Operating System** must be **Linux 64-bit**.
|
|
|
|
.. _cce_10_0141__fig11696366517:
|
|
|
|
.. figure:: /_static/images/en-us_image_0000001518062808.png
|
|
:alt: **Figure 1** Setting parameters
|
|
|
|
**Figure 1** Setting parameters
|
|
|
|
5. After confirming the driver information, click **SEARCH**. A page is displayed, showing the driver information, as shown in :ref:`Figure 2 <cce_10_0141__fig7873421145213>`. Click **DOWNLOAD**.
|
|
|
|
.. _cce_10_0141__fig7873421145213:
|
|
|
|
.. figure:: /_static/images/en-us_image_0000001517743660.png
|
|
:alt: **Figure 2** Driver information
|
|
|
|
**Figure 2** Driver information
|
|
|
|
6. Obtain the driver link in either of the following ways:
|
|
|
|
- Method 1: As shown in :ref:`Figure 3 <cce_10_0141__fig5901194614534>`, find *url=/tesla/470.103.01/NVIDIA-Linux-x86_64-470.103.01.run* in the browser address box. Then, supplement it to obtain the driver link https://us.download.nvidia.com/tesla/470.103.01/NVIDIA-Linux-x86_64-470.103.01.run. By using this method, you must bind an EIP to each GPU node.
|
|
|
|
- Method 2: As shown in :ref:`Figure 3 <cce_10_0141__fig5901194614534>`, click **AGREE & DOWNLOAD** to download the driver. Then, upload the driver to OBS and record the OBS URL. By using this method, you do not need to bind an EIP to GPU nodes.
|
|
|
|
.. _cce_10_0141__fig5901194614534:
|
|
|
|
.. figure:: /_static/images/en-us_image_0000001517903240.png
|
|
:alt: **Figure 3** Obtaining the link
|
|
|
|
**Figure 3** Obtaining the link
|
|
|
|
.. |image1| image:: /_static/images/en-us_image_0000001518062812.png
|