1. Home
  3. Installing NVIDIA DGX software stack in Bright Ubuntu 20.04 software images

Installing NVIDIA DGX software stack in Bright Ubuntu 20.04 software images

This document describes the procedure for installing the official Nvidia DGX software stack in a Bright Ubuntu 20.04 software image. The instructions in this document target the DGX A100, but the same procedure can be used for other DGX systems such as DGX-1 and DGX-2.

Note:  We now publish a pre-made DGX Software image which eliminates the need for these steps.  The following example is only applicable to those who cannot run that software image.  The pre-made image can be obtained from support.  

Step 1: Prepare a copy of the software image

This example uses the default software image as the base image. Any other software image can be used for this purpose.

Clone the base software image to an image called dgxa100-image.

# Create a clone of the default Bright software image
$ cmsh
  clone default-image dgxa100-image

# Important: Remove the Bright cuda-dcgm/cuda-driver packages from software image(if installed)
  cm-chroot-sw-img /cm/images/dgxa100-image
  apt remove cuda-dcgm cuda-driver

Step 2: Assign new image to the node(s) or category

Define a new node. say dgx-a100 and assign the prepared dgx-image to it. After the initial ramdisk has been generated, provision the DGX node.

In the example below we will set the software image for an individual node, but if you have many DGX nodes it makes more sense to create a node category, make your DGX nodes part of this category, and set the software image for the category (which will then be inherited by all nodes in the category).

$ cmsh
  device use dgx-a100;
  set softwareimage dgxa100-image;

  # Wait for ramdisk to be generated

Step 3: Provision DGX node

Provision the DGX node dgx-a100

Step 4: Install DGX software stack

The steps in this section must be performed on the DGX node dgx-a100 provisioned in Step 3. The URLs, names of the repositories and driver versions in this section are subject to change. Please refer to the official DGX Software Stack Install Guide for latest information on DGX repositories when required.

Install DGX tools and Nvidia drivers

# Enable NVIDIA repositories
curl https://repo.download.nvidia.com/baseos/ubuntu/focal/dgx-repo-files.tgz | sudo tar xzf - -C /

# Update internal APT database
apt update

# Recommended: Upgrade all software packages with the latest versions
apt upgrade

# Install DGX system tools and configurations
apt install -y dgx-a100-system-configurations dgx-a100-system-tools dgx-a100-system-tools-extra 

# Disable the ondemand governor to set the governor to performance mode
systemctl disable ondemand

# Recommended: Disable unattended upgrades
apt purge -y unattended-upgrades

# Install latest kernel
apt install -y linux-generic

# Install NVIDIA CUDA driver
apt install -y nvidia-driver-470-server linux-modules-nvidia-470-server-generic libnvidia-nscq-470 nvidia-modprobe nvidia-fabricmanager-470 datacenter-gpu-manager nv-persistence-mode

# Enable required services
systemctl enable nvidia-fabricmanager nvidia-persistenced nvidia-dcgm

# Install Serial over LAN and NVIDIA System Management tool packages:
apt install -y nvidia-ipmisol nvsm

Step 5: Install Mellanox OFED

# Important: Remove ibsim-utils, ibutils and libumad2sim0 packages if installed
apt remove ibsim-utils ibutils libumad2sim0

# Install the MOFED Driver
apt install -y mlnx-ofed-all nvidia-mlnx-ofed-misc 

# Enable and start the openibd service
systemctl enable --now openibd 

If the systemctl command above fails, please check the KB article The Mellanox IB kernel modules fail to load with mlnx-ofed-49, mlnx-ofed-50, mlnx-ofed-51, mlnx-ofed-52 for resolution.

Step 6: Grab image from DGX node

Sync the disk image from the running DGX node onto the software image dgxa100-image on the head node.

$ cmsh
  device use dgx-a100
  grabimage -w

To prevent the system from being suspended after inactivity, the sleep, suspend and hibernate systemd targets in the software image must be masked as follows:

$ cm-chroot-sw-img /cm/images/dgxa100-image
  systemctl mask sleep.target suspend.target hibernate.target

Step 7: Add nvidia_peermem module

The nvidia_peermem module gets installed as part of Step 5. In order to use GPUDirect over Infiniband, the module must be configured to be loaded automatically when the system boots. Doing this by adding it to the kernel modules list for the software image is not recommended, because the nvidia_peermem kernel module relies on other kernel modules that may not be loadable from the initrd. Write the following file in the software image on the head node:

echo "nvidia_peermem" > /cm/images/dgxa100-image/etc/modules-load.d/nvidia_peermem.conf

Step 8: Set kernel version and create initial ramdisk

Installation of OS updates in Step 4 can result in a new kernel getting installed (as part of installing the linux-generic package), and the required kernel modules are built against the new kernel. In such a scenario, the kernelversion setting of the software image must be updated as follows:

$ cmsh
  softwareimage use dgxa100-image
  set kernelversion <TAB> (choose the correct kernel version)

  # Wait for ramdisk to be generated

If no additional kernel was installed, then the ramdisk re-creation must be triggered manually as follows:

$ cmsh
  softwareimage createramdisk dgxa100-image

Step 9: Provision node(s) with new image

The image is now ready to boot any number of DGX nodes. If you have created a Bright node category, you can configure any node to be part of that category by setting its category property. This will make the nodes inherit the softwareimage setting that you have defined for the category. Alternatively, you can configure the softwareimage property for individual nodes.

When nodes are powered on, they will be imaged using the Bright software image to which we have added the DGX software stack. The following command can be used to verify that the Nvidia drivers and services are working as expected:

[root@dgx-a100 ~]# dcgmi discovery -l
8 GPUs found.
| GPU ID | Device Information                                                   |
| 0      | Name: A100-SXM4-40GB                                                 |
|        | PCI Bus ID: 00000000:07:00.0                                         |
|        | Device UUID: GPU-8cda93c7-c7b3-5bb1-c5ae-d18f14ec21b5                |
| 1      | Name: A100-SXM4-40GB                                                 |
|        | PCI Bus ID: 00000000:0F:00.0                                         |
|        | Device UUID: GPU-4b1d7230-8739-451f-4143-19e35cc34e3b                |
| 2      | Name: A100-SXM4-40GB                                                 |
|        | PCI Bus ID: 00000000:47:00.0                                         |
|        | Device UUID: GPU-7d4ef5ea-b2f2-d509-a507-a20f2e656dc7                |
| 3      | Name: A100-SXM4-40GB                                                 |
|        | PCI Bus ID: 00000000:4E:00.0                                         |
|        | Device UUID: GPU-a41e5efb-3df9-9846-8ab7-4adb39f0467f                |
| 4      | Name: A100-SXM4-40GB                                                 |
|        | PCI Bus ID: 00000000:87:00.0                                         |
|        | Device UUID: GPU-900945ac-42c9-b6df-e263-9f1477fab578                |
| 5      | Name: A100-SXM4-40GB                                                 |
|        | PCI Bus ID: 00000000:90:00.0                                         |
|        | Device UUID: GPU-5f5041fd-791b-de04-d7fd-510e4c78de3f                |
| 6      | Name: A100-SXM4-40GB                                                 |
|        | PCI Bus ID: 00000000:B7:00.0                                         |
|        | Device UUID: GPU-a558bb3a-8f95-0487-2b73-46ec419410bf                |
| 7      | Name: A100-SXM4-40GB                                                 |
|        | PCI Bus ID: 00000000:BD:00.0                                         |
|        | Device UUID: GPU-d2839297-b553-cea3-e2bf-c76eff99279b                |
6 NvSwitches found.
| Switch ID |
| 9         |
| 13        |
| 11        |
| 10        |
| 8         |
| 12        |

The node should come up with the correct Mellanox OFED kernel modules and nvidia_peermem kernel module loaded. This can be verified as follows:

root@dgx-a100:~# lsmod | grep nvidia_peermem
nvidia_peermem         16384  0
nvidia              39116800  382 nvidia_uvm,nvidia_peermem,nvidia_modeset
ib_uverbs             131072  2 nvidia_peermem,mlx5_ib


Updated on March 20, 2023

Related Articles

Leave a Comment