Alliance Doc

Using cloud GPUs/fr: Difference between revisions

Page Discussion

Using cloud GPUs/fr (view source)

Revision as of 12:35, 17 April 2020

162 bytes removed ,  4 years ago
Updating to match new version of source page
(Updating to match new version of source page)
(Updating to match new version of source page)
Line 1: Line 1:


<languages />
<languages />
== How to use GPU in cloud VMs ==


This howto describes the steps needed to allocate GPU resources to a virtual machine (VM), installing the necessary drivers as well as simple steps to on what to check to see that the GPU is properly allocated and cannow be used.
This guide describes how to allocate GPU resources to a virtual machine (VM), installing the necessary drivers and checking whether the GPU can be used.


To use a GPU within a VM, the instance needs to be deployed with on for the flavors listed below, to make the GPU available to the Operating System via the PCI bus.
== Supported flavors ==
 
To use a GPU within a VM, the instance needs to be deployed on one of the flavors listed below.  The GPU will be available to the operating system via the PCI bus.


* g2-c24-112gb-500
* g2-c24-112gb-500
Line 11: Line 12:
* g1-c14-56gb
* g1-c14-56gb


== Preparing a Debian 10 Instance ==
== Preparing a Debian 10 instance ==
To use the GPU via the PCI bus, the proprietary Nvidia drivers are required. Due to Debian's policy, the drivers are available from the non-free pool only.


===== <u>Enabling the non-free pool</u> =====
To use the GPU via the PCI bus, the proprietary NVIDIA drivers are required. Due to Debian's policy, the drivers are available from the non-free pool only.
Log in via ssh and add the sources below to ''/etc/apt/sources.list'', if not already in there.
 
===== Enable the non-free pool =====
 
Log in using ssh and add the lines below to ''/etc/apt/sources.list'', if they are not already there.


<pre>
<pre>
Line 23: Line 26:
</pre>
</pre>


===== <u>Installing the Nvidia Driver</u> =====
===== Install the NVIDIA driver =====
The following command will update the apt cache, so that apt will be aware of the new software pool sections, runs an upgrade to update the OS to the latest software versions and installs the kernel headers, the nvidia-driver and the pciutils, which will be required to list the devices connected to the PCI bus.  
 
The following command:
* updates the <code>apt</code> cache, so that <code>apt</code> will be aware of the new software pool sections,
* updates the OS to the latest software versions, and  
* installs kernel headers, an NVIDIA driver, and <code>pciutils</code>, which will be required to list the devices connected to the PCI bus.  


<pre>
<pre>
root@gpu2:~# apt-get update && apt-get -y dist-upgrade && apt-get -y install pciutils linux-headers-`uname -r` linux-headers-amd64 nvidia-driver
root@gpu2:~# apt-get update && apt-get -y dist-upgrade && apt-get -y install pciutils linux-headers-`uname -r` linux-headers-amd64 nvidia-driver
</pre>
</pre>
After the installation has finished and the nvidia has been automatically compiled and loaded, the following steps can be used to verify that everything has been prepared to launch the ''nvidia-persistenced'', which will create the device files and makes the GPU accessible to the user space.
 
If this command finishes successfully, the NVIDIA driver will have been compiled and loaded.


* Check if the GPU is exposed on the PCI bus
* Check if the GPU is exposed on the PCI bus
Line 50: Line 58:
</pre>
</pre>


* Check that the nvidia kernel module is loaded
* Check that the <code>nvidia</code> kernel module is loaded
<pre>
<pre>
root@gpu2:~# lsmod | grep nvidia
root@gpu2:~# lsmod | grep nvidia
Line 57: Line 65:
</pre>
</pre>


Now the userspace process can be started, which will create the necessary character device files.
* Start <code>nvidia-persistenced</code>, which will create the necessary device files and make the GPU accessible in user space.
<pre>
<pre>
root@gpu2:~# service nvidia-persistenced  restart
root@gpu2:~# service nvidia-persistenced  restart
root@gpu2:~# ls -al /dev/nvidia*
root@gpu2:~# ls -al /dev/nvidia*
crw-rw-rw- 1 root root 195,  0 Mar  6 18:55 /dev/nvidia0
crw-rw-rw- 1 root root 195,  0 Mar  6 18:55 /dev/nvidia0
Line 69: Line 76:
The GPU is now available within the user space and can be used.
The GPU is now available within the user space and can be used.


== Preparing a CentOS 7 Instance ==
== Preparing a CentOS 7 instance ==
Nvidia provides repositories for various distributions, therefore the required software can be installed and maintained via these repositories.


To compile the module sources from the nvidia repository, it is necessary to install dkms to automatically build the modules on kernel updates.
NVIDIA provides repositories for various distributions, therefore the required software can be installed and maintained via these repositories.
It ensures that the GPU is still working after OS updates, dkms is provided in the EPEL repository and additionally the kernel headers and the kernel source needs to be installed
 
before the nvidia driver can be set up.
To compile the module sources from the NVIDIA repository, it is necessary to install <code>dkms</code>.
This will automatically build the modules on kernel updates, and therefore ensures that the GPU is still working after any update of the OS.
<code>dkms</code> is provided in the EPEL repository.
Kernel headers and the kernel source need to be installed before the NVIDIA driver can be set up.
 
===== Enable the EPEL repository and install needed software =====


===== <u>Enabling the EPEL repository and install needed software</u> =====
<pre>
<pre>
[root@gpu-centos centos]# yum -y update && reboot
[root@gpu-centos centos]# yum -y update && reboot
Line 82: Line 92:
</pre>
</pre>


===== <u>Adding the NVIDIA repository and install the driver package</u> =====
===== Add the NVIDIA repository and install the driver package =====
Install the YUM repository:
 
Install the <code>yum</code> repository:
<pre>
<pre>
[root@gpu-centos centos]# yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64/cuda-rhel7.repo
[root@gpu-centos centos]# yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64/cuda-rhel7.repo
Line 89: Line 100:
</pre>
</pre>


Nvidia uses its own gpg key to sign its packages, yum will ask to autoimport it.
NVIDIA uses its own GPG key to sign its packages.  <code>yum</code> will ask to autoimport it. Reply "y" for "yes" when prompted.
 
<pre>
<pre>
Retrieving key from http://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64/7fa2af80.pub
Retrieving key from http://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64/7fa2af80.pub
Line 100: Line 110:
</pre>  
</pre>  


After the installation a reboot is required to properly load the module and create the nvidia device files.
After installation, reboot the VM to properly load the module and create the NVIDIA device files.
<pre>
<pre>
[root@gpu-centos ~]# ls -al /dev/nvidia*
[root@gpu-centos ~]# ls -al /dev/nvidia*
FuzzyBot
Bots
38,757

edits

Retrieved from "https://docs.alliancecan.ca/wiki/Special:MobileDiff/82566"