Using cloud vGPUs: Difference between revisions

From Alliance Doc
Jump to navigation Jump to search
(Marked this version for translation)
(spelling and simplification)
Line 2: Line 2:
<translate>
<translate>
<!--T:2-->
<!--T:2-->
This guide describes how to allocate vGPU resources to a virtual machine (VM), installing the necessary drivers and checking whether the vGPU can be used. Repository access as well as access to the vGPUs, is currently only available within [https://arbutus.cloud.computecanada.ca Arbutus Cloud]. Please note that the documentation below only covers the vGPU driver installation, the CUDA toolkit is not pre-installed.
This guide describes how to  
* allocate vGPU resources to a virtual machine (VM),  
* install the necessary drivers and  
* check whether the vGPU can be used.  
Access to repositories as well as to the vGPUs is currently only available within [https://arbutus.cloud.computecanada.ca Arbutus Cloud]. Please note that the documentation below only covers the vGPU driver installation; the CUDA toolkit is not pre-installed.
The [https://developer.nvidia.com/cuda-toolkit-archive CUDA toolkit] can be installed directly from Nvidia or used from [[Accessing_CVMFS|the CVMFS software stack]].
The [https://developer.nvidia.com/cuda-toolkit-archive CUDA toolkit] can be installed directly from Nvidia or used from [[Accessing_CVMFS|the CVMFS software stack]].


Line 8: Line 12:


<!--T:3-->
<!--T:3-->
To use a vGPU within a VM, the instance needs to be deployed on one of the flavors listed below. The vGPU will be available to the operating system via the PCI bus. While finalizing the setup for more vGPU profiles, the only flavor accessible right now is:
To use a vGPU within a VM, the instance needs to be deployed on one of the flavors listed below. The vGPU will be available to the operating system via the PCI bus.


<!--T:4-->
<!--T:4-->
Line 16: Line 20:


<!--T:24-->
<!--T:24-->
Once the VM is available, make sure to update the OS to the latest available software, including the kernel and reboot the VM to have the latest kernel running.
Once the VM is available, make sure to update the OS to the latest available software, including the kernel. Then reboot the VM to have the latest kernel running.
<pre>
<pre>
[root@centos7]# yum -y update && reboot
[root@centos7]# yum -y update && reboot
Line 22: Line 26:


<!--T:25-->
<!--T:25-->
Since the  proprietary nvidia drivers need to be compiled against the running kernel, the package '''dkms''' is required from the [https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm EPEL Repository]
Since the  proprietary Nvidia drivers need to be compiled against the running kernel, the package '''dkms''' is required from the [https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm EPEL Repository]


<!--T:26-->
<!--T:26-->
Line 30: Line 34:


<!--T:27-->
<!--T:27-->
Install the '''Arbutus Cloud''' [http://repo.arbutus.cloud.computecanada.ca/pulp/repos/centos/arbutus-cloud-vgpu-repo.el7.noarch.rpm repository], it also installs the public key the package are signed with to ensure their authenticity, since these drivers and
Install the '''Arbutus Cloud''' [http://repo.arbutus.cloud.computecanada.ca/pulp/repos/centos/arbutus-cloud-vgpu-repo.el7.noarch.rpm repository].
userspace tools are carefully tested first against the infrastructure, before they are made available.  
This also installs the public key the packages are signed with to ensure their authenticity
These drivers and user-space tools are carefully tested against the infrastructure before they are made available.  
<pre>
<pre>
[root@centos7]# yum -y install http://repo.arbutus.cloud.computecanada.ca/pulp/repos/centos/arbutus-cloud-vgpu-repo.el7.noarch.rpm
[root@centos7]# yum -y install http://repo.arbutus.cloud.computecanada.ca/pulp/repos/centos/arbutus-cloud-vgpu-repo.el7.noarch.rpm
Line 37: Line 42:


<!--T:28-->
<!--T:28-->
The last step is to install the '''nvidia vGPU packages'''. The kernel module package 'nvidia-vgpu-kmod', will take a few minutes as it compiles the required kernel modules in the background.
The last step is to install the '''nvidia vGPU packages'''.  
The kernel module package 'nvidia-vgpu-kmod' will take a few minutes as it compiles the required kernel modules in the background.
<pre>
<pre>
[root@centos7]# yum -y install nvidia-vgpu-kmod nvidia-vgpu-gridd nvidia-vgpu-tools
[root@centos7]# yum -y install nvidia-vgpu-kmod nvidia-vgpu-gridd nvidia-vgpu-tools
Line 43: Line 49:


<!--T:29-->
<!--T:29-->
After the successful installation, the vGPU is a now accessible and licensed.
If your installation was successful, the vGPU will be accessible and licensed.
Test by running <code>nvidia-smi</code>:
<pre>
<pre>
[root@centos7]# nvidia-smi         
[root@centos7]# nvidia-smi         
Line 71: Line 78:


<!--T:30-->
<!--T:30-->
To check for the license status as well as other information for the vGPU.
To check for the license status as well as other information about the vGPU:


<pre>
<pre>
Line 130: Line 137:


== Preparation of a VM running CentOS8 == <!--T:6-->
== Preparation of a VM running CentOS8 == <!--T:6-->
Once the VM is available, make sure to update the OS to the latest available software, including the kernel and reboot the VM to have the latest kernel running.
Once the VM is available, make sure to update the OS to the latest available software, including the kernel. Then reboot the VM to have the latest kernel running.
<pre>
<pre>
[root@centos8]# dnf -y update && reboot
[root@centos8]# dnf -y update && reboot
Line 136: Line 143:


<!--T:35-->
<!--T:35-->
Since the proprietary nvidia drivers need to be compiled against the running kernel, the package '''dkms''' is required from the [https://dl.fedoraproject.org/pub/epel/epel-release-latest-8.noarch.rpm EPEL Repository]
Since the proprietary Nvidia drivers need to be compiled against the running kernel, the package '''dkms''' is required from the [https://dl.fedoraproject.org/pub/epel/epel-release-latest-8.noarch.rpm EPEL Repository]


<!--T:36-->
<!--T:36-->
Line 144: Line 151:


<!--T:37-->
<!--T:37-->
Install the '''Arbutus Cloud''' [http://repo.arbutus.cloud.computecanada.ca/pulp/repos/centos/arbutus-cloud-vgpu-repo.el8.noarch.rpm repository], it also installs the public key the package are signed with to ensure their authenticity, since these drivers and
Install the '''Arbutus Cloud''' [http://repo.arbutus.cloud.computecanada.ca/pulp/repos/centos/arbutus-cloud-vgpu-repo.el8.noarch.rpm repository].
userspace tools are carefully tested first against the infrastructure, before they are made available.  
It also installs the public key the packages are signed with to ensure their authenticity.
These drivers and user-space tools are carefully tested against the infrastructure before they are made available.  
<pre>
<pre>
[root@centos8]# dnf -y install http://repo.arbutus.cloud.computecanada.ca/pulp/repos/centos/arbutus-cloud-vgpu-repo.el8.noarch.rpm
[root@centos8]# dnf -y install http://repo.arbutus.cloud.computecanada.ca/pulp/repos/centos/arbutus-cloud-vgpu-repo.el8.noarch.rpm
Line 151: Line 159:


<!--T:38-->
<!--T:38-->
The last step is to install the '''nvidia vGPU packages'''. The kernel module package 'nvidia-vgpu-kmod', will take a few minutes as it compiles the required kernel modules in the background.
The last step is to install the '''nvidia vGPU packages'''.  
The kernel module package 'nvidia-vgpu-kmod' will take a few minutes as it compiles the required kernel modules in the background.
<pre>
<pre>
[root@centos8]# dnf -y install nvidia-vgpu-kmod nvidia-vgpu-gridd nvidia-vgpu-tools
[root@centos8]# dnf -y install nvidia-vgpu-kmod nvidia-vgpu-gridd nvidia-vgpu-tools
Line 157: Line 166:


<!--T:39-->
<!--T:39-->
After the successful installation, the vGPU is a now accessible and licensed.
If your installation was successful, the vGPU will be accessible and licensed.
To check on the status, the same '''nvidia-smi''' commands can be used as seen above for Centos7.
Test by running <code>nvidia-smi</code> as shown above for Centos7.


== Preparation of a VM running Debian10 == <!--T:7-->
== Preparation of a VM running Debian10 == <!--T:7-->
Ensure that the latest packages are installed and the system has been booted the latest stable kernel, as dkms will request the latest one available from the debian repositories.
Ensure that the latest packages are installed and the system has been booted with the latest stable kernel, as '''dkms''' will request the latest one available from the Debian repositories.


<!--T:40-->
<!--T:40-->
Line 169: Line 178:


<!--T:41-->
<!--T:41-->
After a successful reboot, the system should have the latest available kernel running and the repository can be installed, by installing the repo package.
After a successful reboot, the system should have the latest available kernel running and the repository can be installed, by installing the <code>arbutus-cloud-repo</code> package.
This package also contains the gpg key all packages are signed with.
This package also contains the gpg key all packages are signed with.


Line 180: Line 189:


<!--T:43-->
<!--T:43-->
The installation of the package will display a warning, since the key is directly imported (for convenience) via the package's post installation procedure.
The installation of the package will display a warning, since the key is directly imported (for convenience) via the package's post-installation procedure.


<!--T:44-->
<!--T:44-->
Line 190: Line 199:


<!--T:45-->
<!--T:45-->
Update of the local apt cache and installation of the vGPU packages.
Update the local apt cache and install the vGPU packages:


<!--T:46-->
<!--T:46-->
Line 198: Line 207:


<!--T:47-->
<!--T:47-->
After the successful installation, the vGPU is a now accessible and licensed. To check on the status, the same '''nvidia-smi''' commands can be used as seen above for Centos7.
If your installation was successful, the vGPU will be accessible and licensed.
Test by running <code>nvidia-smi</code> as shown above for Centos7.


== Preparation of a VM running Ubuntu20 == <!--T:8-->
== Preparation of a VM running Ubuntu20 == <!--T:8-->
Line 209: Line 219:


<!--T:49-->
<!--T:49-->
After a successful reboot, the system should have the latest avaible kernel running and the repository can be installed, by installing the repo package. This package does also contain the gpg key all packages are signed with.  
After a successful reboot, the system should have the latest available kernel running.
Now the repository can be installed by installing the <code>arbutus-cloud-repo</code> package.
This package also contains the gpg key all packages are signed with.


<!--T:50-->
<!--T:50-->
Line 218: Line 230:


<!--T:51-->
<!--T:51-->
The same warning will be displayed since the signature key is added via post install stage when the package is being installed and can be ignored.
A warning will be displayed since the signature key is added in the post-install stage. The warning can be ignored.
Update of the local apt cache and installation of the vGPU packages.
Update the local apt cache and install the vGPU packages:
<pre>
<pre>
root@ubuntu20:~# apt-get update && apt-get -y install nvidia-vgpu-kmod nvidia-vgpu-tools nvidia-vgpu-gridd
root@ubuntu20:~# apt-get update && apt-get -y install nvidia-vgpu-kmod nvidia-vgpu-tools nvidia-vgpu-gridd
Line 225: Line 237:


<!--T:52-->
<!--T:52-->
After the successful installation, the vGPU is a now accessible and licensed. To check on the status, the same '''nvidia-smi''' commands can be used as seen above for Centos7.
If your installation was successful, the vGPU will be accessible and licensed.
Test by running <code>nvidia-smi</code> as shown above for Centos7.


== Preparation of a VM running Ubuntu18 == <!--T:57-->
== Preparation of a VM running Ubuntu18 == <!--T:57-->
Line 236: Line 249:


<!--T:59-->
<!--T:59-->
After a successful reboot, the system should have the latest avaible kernel running and the repository can be installed, by installing the repo package. This package does also contain the gpg key all packages are signed with.  
After a successful reboot, the system should have the latest available kernel running.
Now the repository can be installed by installing the <code>arbutus-cloud-repo</code> package.
This package also contains the gpg key all packages are signed with.


<!--T:60-->
<!--T:60-->
Line 245: Line 260:


<!--T:61-->
<!--T:61-->
The same warning will be displayed since the signature key is added via post install stage when the package is being installed and can be ignored.
A warning will be displayed since the signature key is added in the post-install stage. The warning can be ignored.
Update of the local apt cache and installation of the vGPU packages.
Update the local apt cache and install the vGPU packages:
<pre>
<pre>
root@ubuntu18:~# apt-get update && apt-get -y install nvidia-vgpu-kmod nvidia-vgpu-tools nvidia-vgpu-gridd
root@ubuntu18:~# apt-get update && apt-get -y install nvidia-vgpu-kmod nvidia-vgpu-tools nvidia-vgpu-gridd
Line 252: Line 267:


<!--T:62-->
<!--T:62-->
After the successful installation, the vGPU is a now accessible and licensed. To check on the status, the same '''nvidia-smi''' commands can be used as seen above for Centos7.
If your installation was successful, the vGPU will be accessible and licensed.
Test by running <code>nvidia-smi</code> as shown above for Centos7.


</translate>
</translate>

Revision as of 19:27, 21 September 2021

Other languages:

This guide describes how to

  • allocate vGPU resources to a virtual machine (VM),
  • install the necessary drivers and
  • check whether the vGPU can be used.

Access to repositories as well as to the vGPUs is currently only available within Arbutus Cloud. Please note that the documentation below only covers the vGPU driver installation; the CUDA toolkit is not pre-installed. The CUDA toolkit can be installed directly from Nvidia or used from the CVMFS software stack.

Supported flavors

To use a vGPU within a VM, the instance needs to be deployed on one of the flavors listed below. The vGPU will be available to the operating system via the PCI bus.

  • g1-8gb-c4-22gb

Preparation of a VM running CentOS7

Once the VM is available, make sure to update the OS to the latest available software, including the kernel. Then reboot the VM to have the latest kernel running. 

[root@centos7]# yum -y update && reboot

Since the proprietary Nvidia drivers need to be compiled against the running kernel, the package dkms is required from the EPEL Repository 

[root@centos7]# yum -y install epel-release

Install the Arbutus Cloud repository. This also installs the public key the packages are signed with to ensure their authenticity These drivers and user-space tools are carefully tested against the infrastructure before they are made available. 

[root@centos7]# yum -y install http://repo.arbutus.cloud.computecanada.ca/pulp/repos/centos/arbutus-cloud-vgpu-repo.el7.noarch.rpm

The last step is to install the nvidia vGPU packages. The kernel module package 'nvidia-vgpu-kmod' will take a few minutes as it compiles the required kernel modules in the background. 

[root@centos7]# yum -y install nvidia-vgpu-kmod nvidia-vgpu-gridd nvidia-vgpu-tools

If your installation was successful, the vGPU will be accessible and licensed. Test by running nvidia-smi: 

[root@centos7]# nvidia-smi         
Tue Sep 21 17:40:33 2021       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 460.91.03    Driver Version: 460.91.03    CUDA Version: 11.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  GRID V100D-8C       On   | 00000000:00:05.0 Off |                  N/A |
| N/A   N/A    P0    N/A /  N/A |    560MiB /  8192MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

To check for the license status as well as other information about the vGPU: 

[root@centos7]# nvidia-smi -q |less
==============NVSMI LOG==============

Timestamp                                 : Tue Sep 21 17:41:48 2021
Driver Version                            : 460.91.03
CUDA Version                              : 11.2

Attached GPUs                             : 1
GPU 00000000:00:05.0
    Product Name                          : GRID V100D-8C
    Product Brand                         : NVIDIA Virtual Compute Server
    Display Mode                          : Enabled
    Display Active                        : Disabled
    Persistence Mode                      : Enabled
    MIG Mode
        Current                           : N/A
        Pending                           : N/A
    Accounting Mode                       : Disabled
    Accounting Mode Buffer Size           : 4000
    Driver Model
        Current                           : N/A
        Pending                           : N/A
    Serial Number                         : N/A
    GPU UUID                              : GPU-c6d5d6c1-1b00-11ec-b031-a89a79e5169c
    Minor Number                          : 0
    VBIOS Version                         : 00.00.00.00.00
    MultiGPU Board                        : No
    Board ID                              : 0x5
    GPU Part Number                       : N/A
    Inforom Version
        Image Version                     : N/A
        OEM Object                        : N/A
        ECC Object                        : N/A
        Power Management Object           : N/A
    GPU Operation Mode
        Current                           : N/A
        Pending                           : N/A
    GPU Virtualization Mode
        Virtualization Mode               : VGPU
        Host VGPU Mode                    : N/A
    vGPU Software Licensed Product
        Product Name                      : NVIDIA Virtual Compute Server
        License Status                    : Licensed
    IBMNPU
        Relaxed Ordering Mode             : N/A
    PCI
        Bus                               : 0x00
        Device                            : 0x05
        Domain                            : 0x0000
        Device Id                         : 0x1DB610DE
        Bus Id                            : 00000000:00:05.0

Preparation of a VM running CentOS8

Once the VM is available, make sure to update the OS to the latest available software, including the kernel. Then reboot the VM to have the latest kernel running. 

[root@centos8]# dnf -y update && reboot

Since the proprietary Nvidia drivers need to be compiled against the running kernel, the package dkms is required from the EPEL Repository 

[root@centos8]# dnf -y install https://dl.fedoraproject.org/pub/epel/epel-release-latest-8.noarch.rpm

Install the Arbutus Cloud repository. It also installs the public key the packages are signed with to ensure their authenticity. These drivers and user-space tools are carefully tested against the infrastructure before they are made available. 

[root@centos8]# dnf -y install http://repo.arbutus.cloud.computecanada.ca/pulp/repos/centos/arbutus-cloud-vgpu-repo.el8.noarch.rpm

The last step is to install the nvidia vGPU packages. The kernel module package 'nvidia-vgpu-kmod' will take a few minutes as it compiles the required kernel modules in the background. 

[root@centos8]# dnf -y install nvidia-vgpu-kmod nvidia-vgpu-gridd nvidia-vgpu-tools

If your installation was successful, the vGPU will be accessible and licensed. Test by running nvidia-smi as shown above for Centos7.

Preparation of a VM running Debian10

Ensure that the latest packages are installed and the system has been booted with the latest stable kernel, as dkms will request the latest one available from the Debian repositories. 

root@debian10:~# apt-get update && apt-get -y dist-upgrade && reboot

After a successful reboot, the system should have the latest available kernel running and the repository can be installed, by installing the arbutus-cloud-repo package. This package also contains the gpg key all packages are signed with. 

root@debian10:~# apt-get -y install gnupg
root@debian10:~# wget http://repo.arbutus.cloud.computecanada.ca/pulp/deb/debian/pool/main/arbutus-cloud-repo_0.1_all.deb
root@debian10:~# dpkg -i arbutus-cloud-repo_0.1_all.deb

The installation of the package will display a warning, since the key is directly imported (for convenience) via the package's post-installation procedure. 

Setting up arbutus-cloud-repo (0.1) ...
Warning: apt-key should not be used in scripts (called from postinst maintainerscript of the package arbutus-cloud-repo)
OK

Update the local apt cache and install the vGPU packages: 

root@debian10:~# apt-get update && apt-get -y install nvidia-vgpu-kmod nvidia-vgpu-tools nvidia-vgpu-gridd

If your installation was successful, the vGPU will be accessible and licensed. Test by running nvidia-smi as shown above for Centos7.

Preparation of a VM running Ubuntu20

Ensure that the OS is up to date and all the latest patches are installed and the latest stable kernel is running. 

root@ubuntu20:~# apt-get update && apt-get -y dist-upgrade && reboot

After a successful reboot, the system should have the latest available kernel running. Now the repository can be installed by installing the arbutus-cloud-repo package. This package also contains the gpg key all packages are signed with. 

root@ubuntu20:~# wget http://repo.arbutus.cloud.computecanada.ca/pulp/deb/ubuntu/pool/main/arbutus-cloud-repo_0.1ubuntu20_all.deb
root@ubuntu20:~# dpkg -i arbutus-cloud-repo_0.1ubuntu20_all.deb

A warning will be displayed since the signature key is added in the post-install stage. The warning can be ignored. Update the local apt cache and install the vGPU packages: 

root@ubuntu20:~# apt-get update && apt-get -y install nvidia-vgpu-kmod nvidia-vgpu-tools nvidia-vgpu-gridd

If your installation was successful, the vGPU will be accessible and licensed. Test by running nvidia-smi as shown above for Centos7.

Preparation of a VM running Ubuntu18

Ensure that the OS is up to date and all the latest patches are installed and the latest stable kernel is running. 

root@ubuntu18:~# apt-get update && apt-get -y dist-upgrade && reboot

After a successful reboot, the system should have the latest available kernel running. Now the repository can be installed by installing the arbutus-cloud-repo package. This package also contains the gpg key all packages are signed with. 

root@ubuntu18:~# wget http://repo.arbutus.cloud.computecanada.ca/pulp/deb/ubuntu18/pool/main/arbutus-cloud-repo_0.1ubuntu18_all.deb
root@ubuntu18:~# dpkg -i arbutus-cloud-repo_0.1ubuntu18_all.deb

A warning will be displayed since the signature key is added in the post-install stage. The warning can be ignored. Update the local apt cache and install the vGPU packages: 

root@ubuntu18:~# apt-get update && apt-get -y install nvidia-vgpu-kmod nvidia-vgpu-tools nvidia-vgpu-gridd

If your installation was successful, the vGPU will be accessible and licensed. Test by running nvidia-smi as shown above for Centos7.

Retrieved from "https://docs.alliancecan.ca/mediawiki/index.php?title=Using_cloud_vGPUs&oldid=103884"