1 of 2

Enable GPU Acceleration

OvenMediaEngine supports GPU-based hardware decoding and encoding. Currently supported GPU acceleration devices are Intel's QuickSync and NVIDIA's NVDECODE/NVENCODE. This document describes how to install the video driver for OvenMediaEngine to use the GPU and how to set the Config file. Please check what graphics card you have and refer to the NVIDIA or Intel driver installation guide.

1. Install GPU Driver

Install Intel QuickSync Driver

If you are using an Intel CPU that supports QuickSync, please refer to the following guide to install the driver. The OSes that support installation using the provided scripts are CentOS 7/8 and Ubuntu 18/20 versions. If you want to install the driver on a different OS, please refer to the Manual Installation Guide document.

When the Intel QuickSync driver installation is complete, the OS must be rebooted for normal operation.

(curl -LOJ https://github.com/AirenSoft/OvenMediaEngine/archive/master.tar.gz && tar xvfz OvenMediaEngine-master.tar.gz) 
OvenMediaEngine-master/misc/install_intel_driver.sh

How to check driver installation

After the driver installation is complete, check whether the driver operates normally with the Matrix Monitor program.

# Use the samples provided in the Intel Media SDK
# Check the list of codecs supported by iGPU
/MediaSDK-intel-mediasdk-21.1.2/build/__bin/release/simple_7_codec

Install NVIDIA GPU Driver

If you are using an NVIDIA graphics card, please refer to the following guide to install the driver. The OS that supports installation with the provided script are CentOS 7/8 and Ubuntu 18/20 versions. If you want to install the driver in another OS, please refer to the manual installation guide document.

CentOS environment requires the process of uninstalling the nouveau driver. After uninstalling the driver, the first reboot is required, and a new NVIDIA driver must be installed and rebooted. Therefore, two install scripts must be executed.

(curl -LOJ https://github.com/AirenSoft/OvenMediaEngine/archive/master.tar.gz && tar xvfz OvenMediaEngine-master.tar.gz)
OvenMediaEngine-master/misc/install_nvidia_driver.sh

How to check driver installation

After the driver installation is complete, check whether the driver is operating normally with the nvidia-smi command.

$ nvidia-smi

Thu Jun 17 10:20:23 2021
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 465.19.01    Driver Version: 465.19.01    CUDA Version: 11.3     |
|-------------------------------+----------------------+----------------------+
| 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  NVIDIA GeForce ...  Off  | 00000000:01:00.0 Off |                  N/A |
| 20%   35C    P8    N/A /  75W |    156MiB /  1997MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|    0   N/A  N/A      1589      G   /usr/libexec/Xorg                  38MiB |
|    0   N/A  N/A      1730      G   /usr/bin/gnome-shell              115MiB |
+-----------------------------------------------------------------------------+

Manual Installation

If the provided installation script fails, please refer to the manual installation guide.

2. Prerequisites

If you have finished installing the driver to use the GPU, you need to reinstall the open source library using Prerequisites.sh . The purpose is to allow external libraries to use the installed graphics driver.

Using Intel QuickSync GPU

OvenMediaEngine-master/misc/prerequisites.sh --enable-qsv

Using NVIDIA GPU

OvenMediaEngine-master/misc/prerequisites.sh --enable-nvc

Using Docker with NVIDIA GPU

Describes how to enable GPU acceleration for users running OvenMediaEngine in the Docker runtime environment. To use GPU acceleration in Docker, the NVIDIA Driver must be installed on the host OS and the NVIDIA Container Toolkit must be installed. This toolkit includes container runtime libraries and utilities to use NVIDIA GPUs in Docker containers.

OvenMediaEngine-master/misc/install_nvidia_docker_container.sh

The NVIDIA Driver must have been previously installed

3. Configuration

To use hardware acceleration, set the HardwareAcceleration option to true under OutputProfiles. If this option is enabled, a hardware codec is automatically used when creating a stream, and if it is unavailable due to insufficient hardware resources, it is replaced with a software codec.

<VirtualHosts>
   <VirtualHost>
      <Name>default</Name>

      <!-- Settings for multi domain and TLS -->
      <Host>
         ...
      </Host>

      <!-- Settings for applications -->
      <Applications>
         <Application>
            <Name>app</Name>
            <Type>live</Type>
            <OutputProfiles>
               <!-- Settings to use hardware codecs -->
               <HardwareAcceleration>true</HardwareAcceleration>
               <OutputProfile>
                  <Name>bypass_stream</Name>
                  <OutputStreamName>${OriginStreamName}_o</OutputStreamName>
                  <Encodes>
                     <Video>
                        <Codec>h264</Codec>
                        <Width>1920</Width>
                        <Height>1080</Height>
                        <Bitrate>5000000</Bitrate>
                        <Framerate>30</Framerate>
                     </Video>
                     
                     <Video>
                        <Codec>h265</Codec>
                        <Width>1280</Width>
                        <Height>720</Height>
                        <Bitrate>5000000</Bitrate>
                        <Framerate>30</Framerate>
                     </Video> 
                  </Encodes>
               </OutputProfile>

            </OutputProfiles>
            <Providers>
               ...
            </Providers>
            <Publishers>
               ...
            </Publishers>
         </Application>
      </Applications>
   </VirtualHost>
</VirtualHosts>

4. Build & Run

You can build the OvenMediaEngine source using the following command. Same as the contents of Getting Started.

sudo apt-get update
cd OvenMediaEngine-master/src
make release
sudo make install
systemctl start ovenmediaengine
# If you want automatically start on boot
systemctl enable ovenmediaengine.service

sudo dnf update
cd OvenMediaEngine-master/src
make release
sudo make install
systemctl start ovenmediaengine
# If you want automatically start on boot
systemctl enable ovenmediaengine.service

sudo yum update
source scl_source enable devtoolset-7
cd OvenMediaEngine-master/src
make release
sudo make install
systemctl start ovenmediaengine
# If you want automatically start on boot
systemctl enable ovenmediaengine.service

How to build and run docker image

To use Docker, you need to build a new Docker image. To build an OvenMediaEngine image with GPU support, you need to set some parameters. First, check the major version of the NVIDIA driver installed on the host OS

Set the major version as the value of the NVIDIA_DRIVER argument when building docker image.

docker build --file Dockerfile -t airensoft/ovenmediaengine:dev --build-arg GPU=TRUE --build-arg NVIDIA_DRIVER=470

After the build is complete, you must include the --gpus all option when running Docker

docker run -d -p 1935:1935 -p 4000-4005:4000-4005/udp -p 3333:3333 -p 3478:3478 -p 8080:8080 -p 9000:9000 -p 9999:9999/udp -p 10006-10010:10006-10010/udp --gpus all airensoft/ovenmediaengine:dev

Support Format

The codecs available using hardware accelerators in OvenMediaEngine are as shown in the table below. Different GPUs support different codecs. If the hardware codec is not available, you should check if your GPU device supports the codec.

Device

H264

H265

VP8

VP9

QuickSync

D / E

NVIDIA

D / E

Docker on NVIDIA Container Toolkit

D / E

D : Decoding, E : Encoding

Reference

Quick Sync Video Format : https://en.wikipedia.org/wiki/Intel_Quick_Sync_Video
NVIDIA NVDEC Video Format : https://en.wikipedia.org/wiki/Nvidia_NVDEC
NVIDIA NVENV Video Format : https://en.wikipedia.org/wiki/Nvidia_NVENC
CUDA Toolkit Installation Guide : https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#introduction
NVIDIA Container Toolkit : https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/arch-overview.html#arch-overview
Quick Sync Video format support: https://en.wikipedia.org/wiki/Intel_Quick_Sync_Video

Manual Installation

Intel QuickSync Driver Installation

Platform Specific Installation

sudo apt install -y libdrm-dev xorg xorg-dev openbox libx11-dev 
sudo apt install -y libgl1-mesa-glx libgl1-mesa-dev

# Centos 7 uses the 2.8.x version of cmake by default. 
# It must be changed to version 3.x or higher.
sudo yum remove -y cmake
sudo yum install -y cmake3
sudo ln -s /usr/bin/cmake3 /usr/bin/cmake

sudo yum install -y libdrm-devel libX11-devel libXi-devel

sudo yum install -y libdrm-devel libX11-devel libXi-devel

Common Installation

LibVA

PREFIX=/opt/ovenmediaengine && \
TEMP_PATH=/tmp && \
LIBVA_VERSION=2.11.0 && \
DIR=${TEMP_PATH}/libva && \
mkdir -p ${DIR} && \
cd ${DIR} && \
curl -sLf https://github.com/intel/libva/archive/refs/tags/${LIBVA_VERSION}.tar.gz | tar -xz --strip-components=1 && \
./autogen.sh --prefix="${PREFIX}" && \
make -j$(nproc) && \
sudo make install && \
rm -rf ${DIR})

GMMLIB

PREFIX=/opt/ovenmediaengine && \
TEMP_PATH=/tmp && \
GMMLIB_VERSION=20.4.1 && \
DIR=${TEMP_PATH}/gmmlib && \
mkdir -p ${DIR} && \
cd ${DIR} && \
curl -sLf https://github.com/intel/gmmlib/archive/refs/tags/intel-gmmlib-${GMMLIB_VERSION}.tar.gz | tar -xz --strip-components=1 && \
mkdir -p ${DIR}/build && \
cd ${DIR}/build && \
cmake -DCMAKE_INSTALL_PREFIX="${PREFIX}" .. && \
make -j$(nproc) && \
sudo make install && \
rm -rf ${DIR}

Intel Media Driver

PREFIX=/opt/ovenmediaengine && \
TEMP_PATH=/tmp && \
INTEL_MEDIA_DRIVER_VERSION=20.4.5 && \
DIR_IMD=${TEMP_PATH}/media-driver && \
mkdir -p ${DIR_IMD} && \
cd ${DIR_IMD} && \
curl -sLf https://github.com/intel/media-driver/archive/refs/tags/intel-media-${INTEL_MEDIA_DRIVER_VERSION}.tar.gz  | tar -xz --strip-components=1 && \
DIR_GMMLIB=${TEMP_PATH}/gmmlib && \
mkdir -p ${DIR_GMMLIB} && \
cd ${DIR_GMMLIB} && \
curl -sLf https://github.com/intel/gmmlib/archive/refs/tags/intel-gmmlib-${GMMLIB_VERSION}.tar.gz | tar -xz --strip-components=1 && \
DIR=${TEMP_PATH}/build && \
mkdir -p ${DIR} && \
cd ${DIR} && \
PKG_CONFIG_PATH=${PREFIX}/lib/pkgconfig:${PKG_CONFIG_PATH} cmake \
    $DIR_IMD \
    -DBUILD_TYPE=release \
    -DBS_DIR_GMMLIB="$DIR_GMMLIB/Source/GmmLib" \
    -DBS_DIR_COMMON=$DIR_GMMLIB/Source/Common \
    -DBS_DIR_INC=$DIR_GMMLIB/Source/inc \
    -DBS_DIR_MEDIA=$DIR_IMD \
    -DCMAKE_INSTALL_PREFIX=${PREFIX} \
    -DCMAKE_INSTALL_LIBDIR=${PREFIX}/lib \
    -DINSTALL_DRIVER_SYSCONF=OFF \
    -DLIBVA_DRIVERS_PATH=${PREFIX}/lib/dri && \
sudo make -j$(nproc) install && \
rm -rf ${DIR} && \
rm -rf ${DIR_IMD} && \
rm -rf ${DIR_GMMLIB}

export LIBVA_DRIVERS_PATH=${PREFIX}/lib
export LIBVA_DRIVER_NAME=iHD

Intel Media SDK

PREFIX=/opt/ovenmediaengine && \
TEMP_PATH=/tmp && \
INTEL_MEDIA_SDK_VERSION=20.5.1 && \
DIR=${TEMP_PATH}/medka-sdk && \
mkdir -p ${DIR} && \
cd ${DIR} && \
curl -sLf https://github.com/Intel-Media-SDK/MediaSDK/archive/refs/tags/intel-mediasdk-${INTEL_MEDIA_SDK_VERSION}.tar.gz  | tar -xz --strip-components=1 && \
mkdir -p ${DIR}/build && \
cd ${DIR}/build && \
PKG_CONFIG_PATH=${PREFIX}/lib/pkgconfig:${PKG_CONFIG_PATH} cmake -DCMAKE_INSTALL_PREFIX="${PREFIX}" .. && \
make -j$(nproc) && \
sudo make install && \
rm -rf ${DIR}

All libraries are installed, the system must be rebooted.

NVIDIA Driver Installation

Platform Specific Installation

# Uninstalling a previously installed NVIDIA Driver
sudo apt-get remove --purge nvidia-*
sudo apt-get -y autoremove
sudo apt-get -y update

# Remove the nouveau driver. If the nouveau driver is in use, the nvidia driver cannot be installed.
USE_NOUVEAU=`sudo lshw -class video | grep nouveau`
if [ ! -z "$USE_NOUVEAU" ]; then

        # Disable nouveau Driver
        echo "blacklist nouveau" >> /etc/modprobe.d/blacklist.conf
        echo "blacklist lbm-nouveau" >> /etc/modprobe.d/blacklist.conf
        echo "options nouveau modeset=0" >> /etc/modprobe.d/blacklist.conf
        echo "alias nouveau off" >> /etc/modprobe.d/blacklist.conf
        echo "alias lbm-nouveau off" >> /etc/modprobe.d/blacklist.conf
        sudo update-initramfs -u
        echo "Using a driver display nouveau.Remove the driver and reboot.Reboot and installation script to rerun the nvidia display the driver to complete the installation."

        sleep 5s
        reboot
fi

# Install Nvidia Driver and Nvidia Toolkit
sudo add-apt-repository ppa:graphics-drivers/ppa
sudo apt -y update
sudo apt-get install -y $(ubuntu-drivers devices | grep recommended | awk '{print $3}')
sudo apt-get install -y nvidia-cuda-toolkit curl make

# Update Kernel
yum -y update
yum -y groupinstall "Development Tools"
yum -y install kernel-devel
yum -y install epel-release
yum -y install dkms curl
echo "Reboot is required to run with a new version of the kernel."

# Remove the nouveau driver. If the nouveau driver is in use, the nvidia driver cannot be installed.
USE_NOUVEAU=`sudo lshw -class video | grep nouveau`
if [ ! -z "$USE_NOUVEAU" ]; then

        # Disable nouveau Driver
        sed "s/GRUB_CMDLINE_LINUX=\"\(.*\)\"/GRUB_CMDLINE_LINUX=\"\1 rd.driver.blacklist=nouveau nouveau.modeset=0\"/" /etc/default/grub
        grub2-mkconfig -o /boot/grub2/grub.cfg
        echo "blacklist nouveau" >> /etc/modprobe.d/blacklist.conf
        mv /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r)-nouveau.img
        dracut /boot/initramfs-$(uname -r).img $(uname -r)

        echo "Using a driver display nouveau. so, remove the driver and reboot. "
        echo "After reboot and installation script to rerun the nvidia display the driver to complete the installation."

        sleep 5s
        reboot
fi

# Install Nvidia Driver
# https://www.nvidia.com/en-us/drivers/unix/
systemctl isolate multi-user.target
wget -N https://us.download.nvidia.com/XFree86/Linux-x86_64/460.84/NVIDIA-Linux-x86_64-460.84.run
sh ./NVIDIA-Linux-x86_64-460.84.run --ui=none --no-questions

# Install Nvidia Toolkit
# https://developer.nvidia.com/cuda-downloads
wget -N https://developer.download.nvidia.com/compute/cuda/11.3.1/local_installers/cuda_11.3.1_465.19.01_linux.run
sh cuda_11.3.1_465.19.01_linux.run --silent

# Configure Envionment Variables
echo "Please add the PATH below to the environment variable."
echo ""
echo "export PATH=${PATH}:/usr/local/cuda/bin/"
echo ""
export PATH=${PATH}:/usr/local/cuda/bin/

dnf update -y
dnf groupinstall -y "Development Tools" 
dnf install -y elfutils-libelf-devel "kernel-devel-uname-r == $(uname -r)"

# Remove the nouveau driver. If the nouveau driver is in use, the nvidia driver cannot be installed.
USE_NOUVEAU=`sudo lshw -class video | grep nouveau`
if [ ! -z "$USE_NOUVEAU" ]; then

        # Disable nouveau Driver
        echo "blacklist nouveau" >> /etc/modprobe.d/blacklist.conf
        mv /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r)-nouveau.img
        dracut /boot/initramfs-$(uname -r).img $(uname -r)

        systemctl set-default multi-user.target
        systemctl get-default

        echo "Using a driver display nouveau. so, remove the driver and reboot. "
        echo "After reboot and installation script to rerun the nvidia display the driver to complete the installation."

        sleep 5s
        reboot
fi

wget -N https://us.download.nvidia.com/XFree86/Linux-x86_64/460.84/NVIDIA-Linux-x86_64-460.84.run
sh ./NVIDIA-Linux-x86_64-460.84.run --ui=none --no-questions

# Install Nvidia Toolkit
# https://developer.nvidia.com/cuda-downloads
wget -N https://developer.download.nvidia.com/compute/cuda/11.3.1/local_installers/cuda_11.3.1_465.19.01_linux.run
sh cuda_11.3.1_465.19.01_linux.run --silent

systemctl set-default graphical.target
systemctl get-default

# Configure Envionment Variables
echo "Please add the PATH below to the environment variable."
echo ""
echo "export PATH=${PATH}:/usr/local/cuda/bin/"
echo ""
export PATH=${PATH}:/usr/local/cuda/bin/

Common Installation

NVCC Headers

PREFIX=/opt/ovenmediaengine && \
TEMP_PATH=/tmp && \
NVCC_HEADERS=11.0.10.1 && \
DIR=${TEMP_PATH}/nvcc-hdr && \
mkdir -p ${DIR} && \
cd ${DIR} && \
curl -sLf https://github.com/FFmpeg/nv-codec-headers/releases/download/n${NVCC_HEADERS}/nv-codec-headers-${NVCC_HEADERS}.tar.gz | tar -xz --strip-components=1 && \
sudo make install && \
rm -rf ${DIR}

NVIDIA Container Toolkit Installation

Platform Specific Installation

# Install Docker 2.x
curl https://get.docker.com && sudo systemctl --now enable docker

# Install NVIDIA Docker Toolkit
distribution=$(. /etc/os-release;echo $ID$VERSION_ID) && \
curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add - && \
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list

sudo apt-get update
sudo apt-get install -y nvidia-docker2

# Restart Docker
sudo systemctl restart docker

# Test
sudo docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi

sudo yum install -y yum-utils
sudo yum-config-manager --add-repo=https://download.docker.com/linux/centos/docker-ce.repo
sudo yum repolist -v
sudo yum install -y https://download.docker.com/linux/centos/7/x86_64/stable/Packages/containerd.io-1.4.3-3.1.el7.x86_64.rpm
sudo yum install docker-ce -y
sudo systemctl --now enable docker

# Install NVIDIA Docker Toolkit
distribution=$(. /etc/os-release;echo $ID$VERSION_ID) && \
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.repo | sudo tee /etc/yum.repos.d/nvidia-docker.repo

if [[ "${OSVERSION}" == "7" ]]; then
    sudo yum clean expire-cache
    sudo yum install -y nvidia-docker2
elif [[ "${OSVERSION}" == "8" ]]; then
    sudo dnf clean expire-cache --refresh
    sudo dnf install -y nvidia-docker2
else
    fail_exit
fi    

# Restart Docker
sudo systemctl restart docker    

# Test
sudo docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi

sudo dnf install -y dnf-utils
sudo dnf config-manager --add-repo=https://download.docker.com/linux/centos/docker-ce.repo
sudo dnf repolist -v
sudo dnf install -y https://download.docker.com/linux/centos/7/x86_64/stable/Packages/containerd.io-1.4.3-3.1.el7.x86_64.rpm
sudo dnf install docker-ce -y
sudo systemctl --now enable docker

# Install NVIDIA Docker Toolkit
distribution=$(. /etc/os-release;echo $ID$VERSION_ID) && \
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.repo | sudo tee /etc/yum.repos.d/nvidia-docker.repo

if [[ "${OSVERSION}" == "7" ]]; then
    sudo yum clean expire-cache
    sudo yum install -y nvidia-docker2
elif [[ "${OSVERSION}" == "8" ]]; then
    sudo dnf clean expire-cache --refresh
    sudo dnf install -y nvidia-docker2
else
    fail_exit
fi    

# Restart Docker
sudo systemctl restart docker    

# Test
sudo docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi

Manual Installation

Intel QuickSync Driver Installation

Platform Specific Installation

sudo apt install -y libdrm-dev xorg xorg-dev openbox libx11-dev 
sudo apt install -y libgl1-mesa-glx libgl1-mesa-dev

# Centos 7 uses the 2.8.x version of cmake by default. 
# It must be changed to version 3.x or higher.
sudo yum remove -y cmake
sudo yum install -y cmake3
sudo ln -s /usr/bin/cmake3 /usr/bin/cmake

sudo yum install -y libdrm-devel libX11-devel libXi-devel

sudo yum install -y libdrm-devel libX11-devel libXi-devel

Common Installation

LibVA

PREFIX=/opt/ovenmediaengine && \
TEMP_PATH=/tmp && \
LIBVA_VERSION=2.11.0 && \
DIR=${TEMP_PATH}/libva && \
mkdir -p ${DIR} && \
cd ${DIR} && \
curl -sLf https://github.com/intel/libva/archive/refs/tags/${LIBVA_VERSION}.tar.gz | tar -xz --strip-components=1 && \
./autogen.sh --prefix="${PREFIX}" && \
make -j$(nproc) && \
sudo make install && \
rm -rf ${DIR})

GMMLIB

PREFIX=/opt/ovenmediaengine && \
TEMP_PATH=/tmp && \
GMMLIB_VERSION=20.4.1 && \
DIR=${TEMP_PATH}/gmmlib && \
mkdir -p ${DIR} && \
cd ${DIR} && \
curl -sLf https://github.com/intel/gmmlib/archive/refs/tags/intel-gmmlib-${GMMLIB_VERSION}.tar.gz | tar -xz --strip-components=1 && \
mkdir -p ${DIR}/build && \
cd ${DIR}/build && \
cmake -DCMAKE_INSTALL_PREFIX="${PREFIX}" .. && \
make -j$(nproc) && \
sudo make install && \
rm -rf ${DIR}

Intel Media Driver

PREFIX=/opt/ovenmediaengine && \
TEMP_PATH=/tmp && \
INTEL_MEDIA_DRIVER_VERSION=20.4.5 && \
DIR_IMD=${TEMP_PATH}/media-driver && \
mkdir -p ${DIR_IMD} && \
cd ${DIR_IMD} && \
curl -sLf https://github.com/intel/media-driver/archive/refs/tags/intel-media-${INTEL_MEDIA_DRIVER_VERSION}.tar.gz  | tar -xz --strip-components=1 && \
DIR_GMMLIB=${TEMP_PATH}/gmmlib && \
mkdir -p ${DIR_GMMLIB} && \
cd ${DIR_GMMLIB} && \
curl -sLf https://github.com/intel/gmmlib/archive/refs/tags/intel-gmmlib-${GMMLIB_VERSION}.tar.gz | tar -xz --strip-components=1 && \
DIR=${TEMP_PATH}/build && \
mkdir -p ${DIR} && \
cd ${DIR} && \
PKG_CONFIG_PATH=${PREFIX}/lib/pkgconfig:${PKG_CONFIG_PATH} cmake \
    $DIR_IMD \
    -DBUILD_TYPE=release \
    -DBS_DIR_GMMLIB="$DIR_GMMLIB/Source/GmmLib" \
    -DBS_DIR_COMMON=$DIR_GMMLIB/Source/Common \
    -DBS_DIR_INC=$DIR_GMMLIB/Source/inc \
    -DBS_DIR_MEDIA=$DIR_IMD \
    -DCMAKE_INSTALL_PREFIX=${PREFIX} \
    -DCMAKE_INSTALL_LIBDIR=${PREFIX}/lib \
    -DINSTALL_DRIVER_SYSCONF=OFF \
    -DLIBVA_DRIVERS_PATH=${PREFIX}/lib/dri && \
sudo make -j$(nproc) install && \
rm -rf ${DIR} && \
rm -rf ${DIR_IMD} && \
rm -rf ${DIR_GMMLIB}

export LIBVA_DRIVERS_PATH=${PREFIX}/lib
export LIBVA_DRIVER_NAME=iHD

Intel Media SDK

PREFIX=/opt/ovenmediaengine && \
TEMP_PATH=/tmp && \
INTEL_MEDIA_SDK_VERSION=20.5.1 && \
DIR=${TEMP_PATH}/medka-sdk && \
mkdir -p ${DIR} && \
cd ${DIR} && \
curl -sLf https://github.com/Intel-Media-SDK/MediaSDK/archive/refs/tags/intel-mediasdk-${INTEL_MEDIA_SDK_VERSION}.tar.gz  | tar -xz --strip-components=1 && \
mkdir -p ${DIR}/build && \
cd ${DIR}/build && \
PKG_CONFIG_PATH=${PREFIX}/lib/pkgconfig:${PKG_CONFIG_PATH} cmake -DCMAKE_INSTALL_PREFIX="${PREFIX}" .. && \
make -j$(nproc) && \
sudo make install && \
rm -rf ${DIR}

All libraries are installed, the system must be rebooted.

NVIDIA Driver Installation

Platform Specific Installation

# Uninstalling a previously installed NVIDIA Driver
sudo apt-get remove --purge nvidia-*
sudo apt-get -y autoremove
sudo apt-get -y update

# Remove the nouveau driver. If the nouveau driver is in use, the nvidia driver cannot be installed.
USE_NOUVEAU=`sudo lshw -class video | grep nouveau`
if [ ! -z "$USE_NOUVEAU" ]; then

        # Disable nouveau Driver
        echo "blacklist nouveau" >> /etc/modprobe.d/blacklist.conf
        echo "blacklist lbm-nouveau" >> /etc/modprobe.d/blacklist.conf
        echo "options nouveau modeset=0" >> /etc/modprobe.d/blacklist.conf
        echo "alias nouveau off" >> /etc/modprobe.d/blacklist.conf
        echo "alias lbm-nouveau off" >> /etc/modprobe.d/blacklist.conf
        sudo update-initramfs -u
        echo "Using a driver display nouveau.Remove the driver and reboot.Reboot and installation script to rerun the nvidia display the driver to complete the installation."

        sleep 5s
        reboot
fi

# Install Nvidia Driver and Nvidia Toolkit
sudo add-apt-repository ppa:graphics-drivers/ppa
sudo apt -y update
sudo apt-get install -y $(ubuntu-drivers devices | grep recommended | awk '{print $3}')
sudo apt-get install -y nvidia-cuda-toolkit curl make

# Update Kernel
yum -y update
yum -y groupinstall "Development Tools"
yum -y install kernel-devel
yum -y install epel-release
yum -y install dkms curl
echo "Reboot is required to run with a new version of the kernel."

# Remove the nouveau driver. If the nouveau driver is in use, the nvidia driver cannot be installed.
USE_NOUVEAU=`sudo lshw -class video | grep nouveau`
if [ ! -z "$USE_NOUVEAU" ]; then

        # Disable nouveau Driver
        sed "s/GRUB_CMDLINE_LINUX=\"\(.*\)\"/GRUB_CMDLINE_LINUX=\"\1 rd.driver.blacklist=nouveau nouveau.modeset=0\"/" /etc/default/grub
        grub2-mkconfig -o /boot/grub2/grub.cfg
        echo "blacklist nouveau" >> /etc/modprobe.d/blacklist.conf
        mv /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r)-nouveau.img
        dracut /boot/initramfs-$(uname -r).img $(uname -r)

        echo "Using a driver display nouveau. so, remove the driver and reboot. "
        echo "After reboot and installation script to rerun the nvidia display the driver to complete the installation."

        sleep 5s
        reboot
fi

# Install Nvidia Driver
# https://www.nvidia.com/en-us/drivers/unix/
systemctl isolate multi-user.target
wget -N https://us.download.nvidia.com/XFree86/Linux-x86_64/460.84/NVIDIA-Linux-x86_64-460.84.run
sh ./NVIDIA-Linux-x86_64-460.84.run --ui=none --no-questions

# Install Nvidia Toolkit
# https://developer.nvidia.com/cuda-downloads
wget -N https://developer.download.nvidia.com/compute/cuda/11.3.1/local_installers/cuda_11.3.1_465.19.01_linux.run
sh cuda_11.3.1_465.19.01_linux.run --silent

# Configure Envionment Variables
echo "Please add the PATH below to the environment variable."
echo ""
echo "export PATH=${PATH}:/usr/local/cuda/bin/"
echo ""
export PATH=${PATH}:/usr/local/cuda/bin/

dnf update -y
dnf groupinstall -y "Development Tools" 
dnf install -y elfutils-libelf-devel "kernel-devel-uname-r == $(uname -r)"

# Remove the nouveau driver. If the nouveau driver is in use, the nvidia driver cannot be installed.
USE_NOUVEAU=`sudo lshw -class video | grep nouveau`
if [ ! -z "$USE_NOUVEAU" ]; then

        # Disable nouveau Driver
        echo "blacklist nouveau" >> /etc/modprobe.d/blacklist.conf
        mv /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r)-nouveau.img
        dracut /boot/initramfs-$(uname -r).img $(uname -r)

        systemctl set-default multi-user.target
        systemctl get-default

        echo "Using a driver display nouveau. so, remove the driver and reboot. "
        echo "After reboot and installation script to rerun the nvidia display the driver to complete the installation."

        sleep 5s
        reboot
fi

wget -N https://us.download.nvidia.com/XFree86/Linux-x86_64/460.84/NVIDIA-Linux-x86_64-460.84.run
sh ./NVIDIA-Linux-x86_64-460.84.run --ui=none --no-questions

# Install Nvidia Toolkit
# https://developer.nvidia.com/cuda-downloads
wget -N https://developer.download.nvidia.com/compute/cuda/11.3.1/local_installers/cuda_11.3.1_465.19.01_linux.run
sh cuda_11.3.1_465.19.01_linux.run --silent

systemctl set-default graphical.target
systemctl get-default

# Configure Envionment Variables
echo "Please add the PATH below to the environment variable."
echo ""
echo "export PATH=${PATH}:/usr/local/cuda/bin/"
echo ""
export PATH=${PATH}:/usr/local/cuda/bin/

Common Installation

NVCC Headers

PREFIX=/opt/ovenmediaengine && \
TEMP_PATH=/tmp && \
NVCC_HEADERS=11.0.10.1 && \
DIR=${TEMP_PATH}/nvcc-hdr && \
mkdir -p ${DIR} && \
cd ${DIR} && \
curl -sLf https://github.com/FFmpeg/nv-codec-headers/releases/download/n${NVCC_HEADERS}/nv-codec-headers-${NVCC_HEADERS}.tar.gz | tar -xz --strip-components=1 && \
sudo make install && \
rm -rf ${DIR}

NVIDIA Container Toolkit Installation

Platform Specific Installation

# Install Docker 2.x
curl https://get.docker.com && sudo systemctl --now enable docker

# Install NVIDIA Docker Toolkit
distribution=$(. /etc/os-release;echo $ID$VERSION_ID) && \
curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add - && \
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list

sudo apt-get update
sudo apt-get install -y nvidia-docker2

# Restart Docker
sudo systemctl restart docker

# Test
sudo docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi

sudo yum install -y yum-utils
sudo yum-config-manager --add-repo=https://download.docker.com/linux/centos/docker-ce.repo
sudo yum repolist -v
sudo yum install -y https://download.docker.com/linux/centos/7/x86_64/stable/Packages/containerd.io-1.4.3-3.1.el7.x86_64.rpm
sudo yum install docker-ce -y
sudo systemctl --now enable docker

# Install NVIDIA Docker Toolkit
distribution=$(. /etc/os-release;echo $ID$VERSION_ID) && \
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.repo | sudo tee /etc/yum.repos.d/nvidia-docker.repo

if [[ "${OSVERSION}" == "7" ]]; then
    sudo yum clean expire-cache
    sudo yum install -y nvidia-docker2
elif [[ "${OSVERSION}" == "8" ]]; then
    sudo dnf clean expire-cache --refresh
    sudo dnf install -y nvidia-docker2
else
    fail_exit
fi    

# Restart Docker
sudo systemctl restart docker    

# Test
sudo docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi

sudo dnf install -y dnf-utils
sudo dnf config-manager --add-repo=https://download.docker.com/linux/centos/docker-ce.repo
sudo dnf repolist -v
sudo dnf install -y https://download.docker.com/linux/centos/7/x86_64/stable/Packages/containerd.io-1.4.3-3.1.el7.x86_64.rpm
sudo dnf install docker-ce -y
sudo systemctl --now enable docker

# Install NVIDIA Docker Toolkit
distribution=$(. /etc/os-release;echo $ID$VERSION_ID) && \
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.repo | sudo tee /etc/yum.repos.d/nvidia-docker.repo

if [[ "${OSVERSION}" == "7" ]]; then
    sudo yum clean expire-cache
    sudo yum install -y nvidia-docker2
elif [[ "${OSVERSION}" == "8" ]]; then
    sudo dnf clean expire-cache --refresh
    sudo dnf install -y nvidia-docker2
else
    fail_exit
fi    

# Restart Docker
sudo systemctl restart docker    

# Test
sudo docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi

Enable GPU Acceleration

1. Install GPU Driver

Install Intel QuickSync Driver

When the Intel QuickSync driver installation is complete, the OS must be rebooted for normal operation.

(curl -LOJ https://github.com/AirenSoft/OvenMediaEngine/archive/master.tar.gz && tar xvfz OvenMediaEngine-master.tar.gz) 
OvenMediaEngine-master/misc/install_intel_driver.sh

How to check driver installation

After the driver installation is complete, check whether the driver operates normally with the Matrix Monitor program.

# Use the samples provided in the Intel Media SDK
# Check the list of codecs supported by iGPU
/MediaSDK-intel-mediasdk-21.1.2/build/__bin/release/simple_7_codec

Install NVIDIA GPU Driver

(curl -LOJ https://github.com/AirenSoft/OvenMediaEngine/archive/master.tar.gz && tar xvfz OvenMediaEngine-master.tar.gz)
OvenMediaEngine-master/misc/install_nvidia_driver.sh

How to check driver installation

After the driver installation is complete, check whether the driver is operating normally with the nvidia-smi command.

$ nvidia-smi

Thu Jun 17 10:20:23 2021
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 465.19.01    Driver Version: 465.19.01    CUDA Version: 11.3     |
|-------------------------------+----------------------+----------------------+
| 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  NVIDIA GeForce ...  Off  | 00000000:01:00.0 Off |                  N/A |
| 20%   35C    P8    N/A /  75W |    156MiB /  1997MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|    0   N/A  N/A      1589      G   /usr/libexec/Xorg                  38MiB |
|    0   N/A  N/A      1730      G   /usr/bin/gnome-shell              115MiB |
+-----------------------------------------------------------------------------+

Manual Installation

If the provided installation script fails, please refer to the manual installation guide.

Manual Installation

2. Prerequisites

Using Intel QuickSync GPU

OvenMediaEngine-master/misc/prerequisites.sh --enable-qsv

Using NVIDIA GPU

OvenMediaEngine-master/misc/prerequisites.sh --enable-nvc

Using Docker with NVIDIA GPU

OvenMediaEngine-master/misc/install_nvidia_docker_container.sh

The NVIDIA Driver must have been previously installed

3. Configuration

<VirtualHosts>
   <VirtualHost>
      <Name>default</Name>

      <!-- Settings for multi domain and TLS -->
      <Host>
         ...
      </Host>

      <!-- Settings for applications -->
      <Applications>
         <Application>
            <Name>app</Name>
            <Type>live</Type>
            <OutputProfiles>
               <!-- Settings to use hardware codecs -->
               <HardwareAcceleration>true</HardwareAcceleration>
               <OutputProfile>
                  <Name>bypass_stream</Name>
                  <OutputStreamName>${OriginStreamName}_o</OutputStreamName>
                  <Encodes>
                     <Video>
                        <Codec>h264</Codec>
                        <Width>1920</Width>
                        <Height>1080</Height>
                        <Bitrate>5000000</Bitrate>
                        <Framerate>30</Framerate>
                     </Video>
                     
                     <Video>
                        <Codec>h265</Codec>
                        <Width>1280</Width>
                        <Height>720</Height>
                        <Bitrate>5000000</Bitrate>
                        <Framerate>30</Framerate>
                     </Video> 
                  </Encodes>
               </OutputProfile>

            </OutputProfiles>
            <Providers>
               ...
            </Providers>
            <Publishers>
               ...
            </Publishers>
         </Application>
      </Applications>
   </VirtualHost>
</VirtualHosts>

Configuration

4. Build & Run

You can build the OvenMediaEngine source using the following command. Same as the contents of Getting Started.

sudo apt-get update
cd OvenMediaEngine-master/src
make release
sudo make install
systemctl start ovenmediaengine
# If you want automatically start on boot
systemctl enable ovenmediaengine.service

sudo dnf update
cd OvenMediaEngine-master/src
make release
sudo make install
systemctl start ovenmediaengine
# If you want automatically start on boot
systemctl enable ovenmediaengine.service

sudo yum update
source scl_source enable devtoolset-7
cd OvenMediaEngine-master/src
make release
sudo make install
systemctl start ovenmediaengine
# If you want automatically start on boot
systemctl enable ovenmediaengine.service

Getting Started

How to build and run docker image

$ nvidia-smi --query-gpu=driver_version --format=csv
driver_version

Set the major version as the value of the NVIDIA_DRIVER argument when building docker image.

docker build --file Dockerfile -t airensoft/ovenmediaengine:dev --build-arg GPU=TRUE --build-arg NVIDIA_DRIVER=470

After the build is complete, you must include the --gpus all option when running Docker

docker run -d -p 1935:1935 -p 4000-4005:4000-4005/udp -p 3333:3333 -p 3478:3478 -p 8080:8080 -p 9000:9000 -p 9999:9999/udp -p 10006-10010:10006-10010/udp --gpus all airensoft/ovenmediaengine:dev

Support Format

Device

H264

H265

VP8

VP9

QuickSync

D / E

NVIDIA

D / E

Docker on NVIDIA Container Toolkit

D / E

D : Decoding, E : Encoding

Reference

Quick Sync Video Format : https://en.wikipedia.org/wiki/Intel_Quick_Sync_Video
NVIDIA NVDEC Video Format : https://en.wikipedia.org/wiki/Nvidia_NVDEC
NVIDIA NVENV Video Format : https://en.wikipedia.org/wiki/Nvidia_NVENC
CUDA Toolkit Installation Guide : https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#introduction
NVIDIA Container Toolkit : https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/arch-overview.html#arch-overview
Quick Sync Video format support: https://en.wikipedia.org/wiki/Intel_Quick_Sync_Video