intel/media-delivery

This collection of samples demonstrates best practices to achieve optimal video quality and performance on Intel GPUs for content delivery networks. Check out our demo, recommended command lines and quality and performance measuring tools.

Media Delivery Software Stack

Contents

• Background
• Overview
• Prerequisites
  • To run Intel® Data Center GPU Flex Series
  • To run Intel® Arc™ A-Series Graphics
  • To run upstreamed GPU products
• Installation
  • Setup Docker
  • Setup Media Delivery
• Running
  • Running 8K with Intel® Deep Link Hyper Encode
• Contributing
• Content Attribution

Background

Intel GPUs have dedicated hardware that enables fast, energy-efficient encoding and decoding of video compliant with the industry standards such as H.264, HEVC and AV1. Drivers and software enable application developers to utilize this hardware either directly from their own applications or from within popular frameworks such as FFmpeg and GStreamer.

Overview

This repository curates a coherent set of software, drivers and utilities to assist users who are getting started with transcoding video on Intel GPUs so that they can begin evaluating performance and incorporating Intel GPU transcode acceleration into their applications.

Included here are scripts, Docker configurations and documents that provide:

• Kernel mode drivers.
  • The latest Intel GPUs are not fully supported by currently available Linux distributions so backported drivers can be installed by following the instructions below.
• User mode drivers and libraries:
  • Intel® Media Driver
  • Intel® oneAPI Video Processing Library (oneVPL)
  • Intel® Media SDK
• Transcoding software:
  • FFmpeg with ffmpeg-qsv plugins (qsv means "QuickSync", a name for Intel's GPU video transcoding technology)
• Quality and performance measuring software:
  • Set of quality and performance benchmarking scripts
  • VMAF
  • Intel GPU Tools
• Demonstrations:
  • Samples which demonstrate operations typical for Content Delivery Network (CDN) applications such as Video On Demand (VOD) streaming under Nginx server.
• Setup guides and user documentation:
  • GPU virtualization setup guide
  • Capabilities of ffmpeg-qsv plugins
• Benchmarking data for specific GPUs:
  • Benchmarks for Intel® Data Center GPU Flex Series
  • Benchmarks for Intel® Iris® Xe MAX graphics

Prerequisites

To run Intel® Data Center GPU Flex Series

We recommend to attach GPU of Intel® Data Center GPU Flex Series to the system with the following characteristics:

	Recommendation
CPU	3rd Generation Intel® Xeon® Scalable Processors
Storage	At least of 20GB of free disk space available for docker
Operating System(s)	Ubuntu 20.04 or Ubuntu 22.04 with kernel mode support as described here

To run Intel® Arc™ A-Series Graphics

GPU of Intel® Arc™ A-Series Graphics should be attached to the system with the following requirements:

	Recommendation
CPU	One of the following CPUs: 12th Generation Intel® Core™ i9 Processors 12th Generation Intel® Core™ i7 Processors 12th Generation Intel® Core™ i5 Processors 12th Generation Intel® Core™ i3 Processors 11th Generation Intel® Core™ i9 Processors 11th Generation Intel® Core™ i7 Processors 11th Generation Intel® Core™ i5 Processors 11th Generation Intel® Core™ i3 Processors
Storage	At least of 20GB of free disk space available for docker
Operating System(s)	Ubuntu 20.04 or Ubuntu 22.04 with kernel mode support as described here

To run upstreamed GPU products

To run upstream GPU products, i.e. those GPUs support for which is available in Linux distributions, you need:

	Recommendation
Operating System(s)	Linux distribution capable to support your GPU. We recommend Ubuntu 20.04 or later.
Storage	At least of 20GB of free disk space available for docker

Installation

We provide dockerfiles to setup transcoding environment. These dockerfiles need to be built into docker containers. Dockerfiles can also be used as a reference instruction to install ingredients on bare metal.

If you want to run in virtual environment, first follow instruction given in GPU virtualization setup guide and then proceed with the setup described below.

Setup Docker

Docker is required to build and to run media delivery containers. If you run Ubuntu 20.04 or later you can install it as follows:

sudo apt-get install docker.io

You might need to further configure docker on your system:

• Allow docker to run under your user account (remember to relogin for group modification to take effect):

  sudo usermod -aG docker $(whoami) && exit
  

• Consider to register and login to Docker Hub. Docker Hub limits the number of docker image downloads ("pulls"). For anonymous users this limit is tied to IP address. For authenticated users it depends on subscription type. For details see https://docs.docker.com/docker-hub/download-rate-limit/. To authenticate running docker engine, execute:

  # you will be prompted to enter username and password:
  docker login
  

• If you run behind a proxy, configure proxies for for docker daemon. Refer to https://docs.docker.com/config/daemon/systemd/. Below example assumes that you have http_proxy environment variable set in advance:

  sudo mkdir -p /etc/systemd/system/docker.service.d
  echo "[Service]" | sudo tee /etc/systemd/system/docker.service.d/https-proxy.conf
  echo "Environment=\"HTTPS_PROXY=$http_proxy/\"" | \
    sudo tee -a /etc/systemd/system/docker.service.d/https-proxy.conf
  
  sudo systemctl daemon-reload
  sudo systemctl restart docker
  
  sudo systemctl show --property=Environment --no-pager docker
  

• Make sure that docker has at least 20GB of hard disk space to use. To check available space run (in the example below 39GB are available):

  $ df -h $(docker info -f '{{ .DockerRootDir}}')
  Filesystem      Size  Used Avail Use% Mounted on
  /dev/sda1        74G   32G   39G  46% /
  

  If disk space is not enough (for example, default /var/lib/docker is mounted to a small size partition which might be a case for /var), consider reconfiguring docker storage location as follows:

  # Below assumes unaltered default docker installation when
  # /etc/docker/daemon.json does not exist
  echo "{\"data-root\": \"/mnt/newlocation\"}" | sudo tee /etc/docker/daemon.json
  sudo systemctl daemon-reload
  sudo systemctl restart docker
  

Setup Media Delivery

We provide few different setup configurations which differ by versions and origins of the included Intel media stack components. Some versions of media stack require special setup for the host.

Dockerfile	Intel media stack origin	Supported Intel GPUs	Host setup instructions
docker/ubuntu20.04/selfbuild-prodkmd/Dockerfile	Self-built from open source	Alchemist, ATS-M	Intel GPU DKMS
docker/ubuntu20.04/selfbuild/Dockerfile	Self-built from open source	Gen8+ (legacy upstreamed platforms), such as SKL, KBL, CFL, TGL, DG1, etc.	Use any Linux distribution which supports required platform
docker/ubuntu20.04/native/Dockerfile	Ubuntu 20.04	Gen8+, check Ubuntu 20.04 documentation	Use any Linux distribution which supports required platform

To build any of the configurations, first clone Media Delivery repository:

git clone https://github.com/intel/media-delivery.git && cd media-delivery

To build configuration which targets DG2/ATS-M stack self-built from open source projects, run:

docker build \
  $(env | grep -E '(_proxy=|_PROXY)' | sed 's/^/--build-arg /') \
  --file docker/ubuntu20.04/selfbuild-prodkmd/Dockerfile \
  --tag intel-media-delivery .

To build configuration which targets Gen8+ legacy upstreamed platforms via stack self-built from open source projects, run:

docker build \
  $(env | grep -E '(_proxy=|_PROXY)' | sed 's/^/--build-arg /') \
  --file docker/ubuntu20.04/selfbuild/Dockerfile \
  --tag intel-media-delivery .

Running

Docker containers provide isolated environments with configured software. To access resources on a host system you need to add specific options when starting docker containers. Overall, software included into media-delivery constainers requires the following:

• To access desired GPU you need to map it to the container, see --device option below
• To be able to access performance metrics, you need --cap-add SYS_ADMIN
• To access ngingx server (if you are running a demo), you need to forward 8080 port, see -p 8080:8080

Summarizing, start container as follows (-v option maps a host folder to the container so you can copy transcoded streams back to the host):

DEVICE=${DEVICE:-/dev/dri/renderD128}
DEVICE_GRP=$(stat --format %g $DEVICE)
mkdir -p /tmp/media-delivery && chmod -R 777 /tmp/media-delivery
docker run --rm -it -v /tmp/media-delivery:/opt/media-delivery \
  -e DEVICE=$DEVICE --device $DEVICE --group-add $DEVICE_GRP \
  --cap-add SYS_ADMIN \
  -p 8080:8080 \
  intel-media-delivery

Once inside a container you can run the included software and scripts. To start, we recommend running simple scripts which will showcase basic transcoding capabilities. These scripts will download sample video clips, though you can supply your own as a script argument if needed. If you work under proxy do not forget to add it to your environment (via export https_proxy=<...>).

• Below commands will run single transcoding session (1080p or 4K) and produce output files which you can copy to the host and review:

  # AV1 to AV1:
  ffmpeg-qsv-AV1-1080p.sh 1
  ffmpeg-qsv-AV1-4K.sh 1
  sample-multi-transcode-AV1-1080p.sh 1
  sample-multi-transcode-AV1-4K.sh 1
  
  # AVC to AVC:
  ffmpeg-qsv-AVC-1080p.sh 1
  ffmpeg-qsv-AVC-4K.sh 1
  sample-multi-transcode-AVC-1080p.sh 1
  sample-multi-transcode-AVC-4K.sh 1
  
  # HEVC to HEVC
  ffmpeg-qsv-HEVC-1080p.sh 1
  ffmpeg-qsv-HEVC-4K.sh 1
  sample-multi-transcode-HEVC-1080p.sh 1
  sample-multi-transcode-HEVC-4K.sh 1
  

• Below commands will run specified number of parallel transcoding sessions (1080p or 4K). No output files will be produced, but you can check performance. Mind that below numbers of parallel transcoding sessions are suggested for Intel® Data Center GPU Flex Series. Other GPUs might support different number of sessions running at realtime:

  # AV1 to AV1:
  ffmpeg-qsv-AV1-1080p.sh 16
  ffmpeg-qsv-AV1-4K.sh 4
  sample-multi-transcode-AV1-1080p.sh 16
  sample-multi-transcode-AV1-4K.sh 4
  
  # AVC to AVC:
  ffmpeg-qsv-AVC-1080p.sh 12
  ffmpeg-qsv-AVC-4K.sh 2
  sample-multi-transcode-AVC-1080p.sh 12
  sample-multi-transcode-AVC-4K.sh 2
  
  # HEVC to HEVC
  ffmpeg-qsv-HEVC-1080p.sh 16
  ffmpeg-qsv-HEVC-4K.sh 4
  sample-multi-transcode-HEVC-1080p.sh 16
  sample-multi-transcode-HEVC-4K.sh 4
  

These scripts run transcoding command lines which we recommend to use for best performance and quality in case of Random Access encoding. See reference command lines for details.

In addition to the simple scripts described above, this project provides the following scripts and software which can be tried next:

• Quality and performance benchmark scripts
• CDN Demo

For the more complex samples, check out Open Visual Cloud and their full scale CDN Transcode Sample.

Running 8K with Intel® Deep Link Hyper Encode

Intel® Deep Link Hyper Encode Technology was designed to boost transcode performance to achieve 8K60 real-time throughput. Currently supported via Sample Multi Transcode application, this technology was tested on Intel® Data Center GPU Flex 140 card (2 GPU nodes). Media delivery container can be used to check two different flavors of this approach:

• 1 GPU node solution (encoder and decoder workloads are shared between 2 VDBOX engines of a single GPU node and encoding is parallelized on a GOP level)
• 2 GPU node solution (encoder and decoder workloads use 4 VDBOX engines of 2 GPU nodes and encoding is parallelized on a GOP level)

Simple example scripts are provided allowing for a quick test of 8k60 transcoding at the user’s end. First, start docker with mapping multiple GPU nodes (mind --device /dev/dri vs --device $DEVICE as in previous examples) as follows:

DEVICE=${DEVICE:-/dev/dri/renderD128}
DEVICE_GRP=$(stat --format %g $DEVICE)
mkdir -p /tmp/media-delivery && chmod -R 777 /tmp/media-delivery
docker run --rm -it -v /tmp/media-delivery:/opt/media-delivery \
  -e DEVICE=$DEVICE --device /dev/dri --group-add $DEVICE_GRP \
  --cap-add SYS_ADMIN \
  -p 8080:8080 \
  intel-media-delivery

Once inside the container, simple scripts can be used to showcase Intel® Deep Link Hyper Encode Technology. If you work behind a firewall, please add HTTPS proxy to your environment (via export https_proxy=<...>) before running the scripts.

• Use the following commands to run single 8K transcoding session using either of the two flavors:

  # AV1 to AV1:
  sample-multi-transcode-AV1-8K-hyperenc1gpu.sh
  sample-multi-transcode-AV1-8K-hyperenc2gpu.sh
  
  # HEVC to HEVC
  sample-multi-transcode-HEVC-8K-hyperenc1gpu.sh
  sample-multi-transcode-HEVC-8K-hyperenc2gpu.sh
  

More information on Intel® Deep Link Hyper Encode Technology can be found here:

• Accelerating Media Delivery with Intel® Data Center GPU Flex Series

Contributing

Feedback and contributions are welcome. Please, submit issues and pull requests here at GitHub.

Dockerfiles should be supported as described in the document.

If changes are done to dockerfiles and/or scipts, please, make sure to run tests before submitting pull requests.

Content Attribution

Container image comes with some embedded content attributed as follows:

/opt/data/embedded/WAR_TRAILER_HiQ_10_withAudio.mp4:
  Film: WAR - Courtesy & Copyright: Yash Raj Films Pvt. Ltd.

Inside the container, please, refer to the following file:

cat /opt/data/embedded/usage.txt

media-delivery's Issues

demo abr test fails since abr mode is currently not supported

media-delivery/tests/demo.bats

Line 252 in f2b1428

@test "demo vod/abr from noaudio" {

This test case fails since mode is not currently supported. This is regression caused by 4878006.

intel_gpu_top: error while loading shared libraries: libglib-2.0.so.0 (for selfbuild docker)

$ docker run -it --rm --device=/dev/dri/renderD128 --group-add=109 --cap-add SYS_ADMIN intel-media-delivery-selfbuild

user@a10c2b58909c:~$ intel_gpu_top
intel_gpu_top: error while loading shared libraries: libglib-2.0.so.0: cannot open shared object file: No such file or directory
```

Recommend PERFMON capability instead of SYS_ADMIN

In: https://github.com/intel/media-delivery/blob/master/doc/howto.rst

You state:
"Such metrics require CAP_SYS_ADMIN or /proc/sys/kernel/perf_event_paranoid<=0 to be collected."

I would would suggest replacing that with:
"Such metrics require PERFMON capability or /proc/sys/kernel/perf_event_paranoid<=0 to be collected. On older kernels and container runtime versions which are lacking support for the PERFMON capability, SYS_ADMIN capability needs to be used instead."

(Because SYS_ADMIN capability grants too wide range of privileges, kernel/security folks are replacing its use with finer-grained capabilities.)

As long PERFMON is mentioned, I think it's fine to use CAP_SYS_ADMIN in your usage examples, because there are still a lot of setups that do not support PERFMON capability yet.

measure-quality: support transcoding w/ mediasdk smt

https://github.com/intel/media-delivery#known-limitations

measure-quality does not support transcoding of input streams in raw video formats w/ Media SDK sample applications (i.e. you will get results for ffmpeg-qsv only)

Need to remove this limitation.

Fix shellcheck warnings in the scripts

Pre-condition:

• sudo apt install shellcheck or dnf install shellcheck

There are quite a few shell scripts:

find -name '*.*sh' | wc -l
38

And they trigger a lot of shellcheck warnings:

find -name '*.*sh' | xargs shellcheck | grep SC[0-9]*: | wc -l
658

---

Note: one should not follow shellcheck recommendations blindly. Main one where one could screw up is changing this:

base_opts="-a -b -c"
command $base_opts -d -e -f

To:

base_opts="-a -b -c"
command "$base_opts" -d -e -f

My preferred way is just adding shellcheck override for them, or switching to helper function like this:

run_command () {
   command -a -b -c $*
}
run_command -d -e -f

Whichever makes more sense for given code.

DKMS issue in GAUDI 1

Hi ALL

I currently using Gaudi 1 with below information:
kernel version: 5.15.0-73-generic.
Currently driver version is 1.17.0-495.
Firmware version is 1.10.0-494.
OS : Distributor ID: Ubuntu
Description: Ubuntu 22.04.4 LTS
Release: 22.04
Codename: jammy

Now I faced the dkms package cannot installed and habanalab driver is not loaded or AIP aborting issue.
Hope can hear back for this issue to make the server run.

measure/perf does not support streams w/ floating point framerate

measure-perf does not correctly handle streams w/ floating point framerates. For example, consider stream w/ framerate=29.97:

#####################################################################
PNP MEDIA SMT AVC-AVC: MULTISTREAM: 1 & ITERATION: 0
#####################################################################
 CONCURRENT: h264_HP@L4_0 123.58 fps/stream , Meets ContentFPS 28500.0 : False , Within margin 2%: 0 True

measure-perf won't go above 1st iteration and thinks that stream is of crazy framerate=28500.

Compare w/ stream of framerate=25:

#####################################################################
PNP MEDIA SMT AVC-AVC: MULTISTREAM: 1 & ITERATION: 0
#####################################################################
 CONCURRENT: 1920x1088_fld_0 25.02 fps/stream , Meets ContentFPS 23.75 : True , Within margin 2%: 0 True

measure-quality: clarify behavior with --nframes 0

https://github.com/intel/media-delivery/blob/master/doc/man/measure-quality.asciidoc#processing-options

--nframes|-n <uint>
Process (encode, calculate metrics) this number of frames and stop.

Clarify which behavior to expect with --nframes 0.

measure/perf: support streams in container formats

measure-perf does not support input streams in container formats (i.e. .mp4, .ts files won't be accepted). Need to remove this limitation.

measure/perf: support input streams with any resolution

measure-perf supports input streams with specific resolutions only: 720p, 1080p, 2160p. Need to remove this limitation.

switch to libvmaf-1.5.2

libvmaf-1.5.2 was recently released: https://github.com/Netflix/vmaf/releases/tag/v1.5.2

measure-quality: better handling bd-rate outliers

Tests revealed that one of the test sequences (Netflix's DinnerScene) causes perceptual objective metrics (VMAF/SSIM/MS-SSIM) based BD-rates to max out for HEVC on Gen 9, as x265-veryslow reference (tuned for PSNR) has a significant negative shift for this outlier sequence with film-grain noise. The VMAF tool based BD-rate calculator is maxing out the BD-rate at +-1M% which causes the BD-rate test set average to be significantly skewed. This occurrence needs to be handled better. A solution could be to assign maxed out BD-rate values in the results an "N/A" label and skip it in calculating BD-rate average.

measure/perf fails on mp4 streams

media-delivery/tests/measure-perf.bats

Line 265 in 45999e5

@test "measure perf --skip-perf raw mp4" {

Started after 45999e5.

dkms: cse module can't be built due to missing commit

media-delivery/templates/m4docker/components/cse-backports.m4

Line 33 in 45456c9

DECLARE(`CSE_BACKPORTS_VER',c334b88)

[dvrogozh@dvrscl intel-gpu-cse-backports]$ git show c334b88
fatal: ambiguous argument 'c334b88': unknown revision or path not in the working tree.
Use '--' to separate paths from revisions, like this:
'git <command> [<revision>...] -- [<file>...]'

CSE sources were forced pushed to fix some issue. I think that with the updated checksums this was the commit we've integrated into media-delivery: intel-gpu/intel-gpu-cse-backports@56c6bad. I will check this out on Monday and provide a fix.

demo: support input file w/o audio

https://github.com/intel/media-delivery/blob/master/samples/cdn/do-transcode.sh#L147

We actually account on .mp4 content for the demo to contain audio. This might not be the case if engineers are playing with the quality selected content which usually are pure raw video streams. This, this makes sense to handle such streams.

Support Intel repository channels (https://dgpu-docs.intel.com)

Support Intel repository channels described in https://dgpu-docs.intel.com/installation-guides/ubuntu/ubuntu-focal.html

ci: support github actions

Need to add ci into the project.

Docker CE: No CAP_SYS_ADMIN capability reported while it was set

Specific for Docker CE, works under Docker IO:

$ docker --version
Docker version 20.10.17, build 100c701
$ sudo docker run -it --rm --device=/dev/dri/renderD128 --group-add=109 --cap-add SYS_ADMIN intel-media-delivery:latest capsh --print
error: no CAP_SYS_ADMIN capability (required to get GPU% metrics)
error:   if you run under docker, rerun with:
error:     --cap-add SYS_ADMIN
error: failed to setup demo

measure-perf/smt: segfault on perf. measurement via mediasdk samples on HEVC encoding

$ docker build \
  $(env | grep -E '(_proxy=|_PROXY)' | sed 's/^/--build-arg /') \
  --file Dockerfile.ubuntu \
  --tag intel-media-delivery \
  .
$ docker run  -it --rm --device=/dev/dri/renderD128 --group-add=39 --cap-add SYS_ADMIN intel-media-delivery .

$ ffmpeg -an -hwaccel qsv -qsv_device /dev/dri/renderD128 \
  -c:v h264_qsv -i /opt/data/embedded/WAR_TRAILER_HiQ_10_withAudio.mp4 \
  -vf scale_qsv=w=-1:h=2160,vpp_qsv=framerate=60 \
  -c:v hevc_qsv -preset medium -profile:v main -b:v 1000000 -vframes 20 \
  -y /tmp/WAR.hevc

$ measure perf --verbose --skip-perf /tmp/WAR.hevc
#####################################################################
MSPERF (MULTI STREAMS PERFORMANCE) v0.20.06.03
#####################################################################

ACCEPTED !!! to continue with [utilization/metric/statistic] profiles


#####################################################################
 Profiling: WAR.hevc
 PASS: via FFMPEG/FFPROBE
 content_fps = 60.0 , content_height = 2160 , content_codec = hevc



#####################################################################
PNP MEDIA SMT HEVC-AVC: MULTISTREAM: 1 & ITERATION: 0 - WAR
#####################################################################
 [VERBOSE][CMD] /usr/share/mfx/samples/sample_multi_transcode -fps 60 -i::h265 /tmp/WAR.hevc -hw -async 1 -u 4 -gop_size 256 -dist 8 -num_ref 5 -vbr -b 9000 -NalHrdConformance:off -VuiNalHrdParameters:off -hrd 5000 -InitialDelayInKB 2500 -extbrc::implicit -ExtBrcAdaptiveLTR:on -o::h264 /opt/data/artifacts/measure/perf/output/WAR_hevc_0.h264  -device /dev/dri/renderD128 -p /tmp/perf/WAR_hevc_0_transcode_log.txt >> /tmp/perf/console_log.txt 2>&1

 [VERBOSE][LINUX_PERF_TOOLS] /opt/intel/samples/bin/performance/MSGo.py

 Exit early, hang/error in submitted commands: 239

NOTE: AVC looks working fine. < this was wrong, I mixed with different input stream, this seems input dependent

ffmpeg: can't select device on multi gpu systems

Current way for device selection (w/ -hwaccel qsv -hwaccel_device ) does not actually works. See https://github.com/Intel-Media-SDK/MediaSDK/wiki/FFmpeg-QSV-Multi-GPU-Selection-on-Linux for proper device selection break down.

build: project fails to build due to bumped ffmpeg requirement over vmaf (>=1.5.2)

media-delivery/Dockerfile.ubuntu

Line 162 in b089197

--enable-libvmaf \

With the FFMPEG_VERSION=master which we have above:

media-delivery/Dockerfile.ubuntu

Line 141 in b089197

ARG FFMPEG_VERSION=master

project fails to build starting from this ffmpeg commit: FFmpeg/FFmpeg@4979afd which starts requiring libvmaf>=1.5.2

ffmpeg-hls-client demo hitting 502 bad gateway on multiple streams

ffmpeg-hls-client hitting bad gateway on >1 stream in demo. This is the to-left client pane - sometimes 2/4 streams will run, sometimes 1/4:

Every 1.0s: bash -c watch_pids 3712d5e110d9: Tue Mar 28 05:32:22 2023

ffmpeg streaming clients monitor

Output and logs path: /opt/data/artifacts/ffmpeg-hls-client
Total clients: 4
Running clients: 2
vod/avc/WAR_TRAILER_HiQ_10_withAudio-1: size=21M, frames=1023, fps=24
vod/avc/WAR_TRAILER_HiQ_10_withAudio-2: size=21M, frames=1023, fps=24
Completed clients: 2
vod/avc/WAR_TRAILER_HiQ_10_withAudio-3: size=, frames=, fps=, status=1
vod/avc/WAR_TRAILER_HiQ_10_withAudio-4: size=, frames=, fps=, status=1

CTRL^C to exit monitor and enter shell

The logs in /opt/data/artifacts/ffmpeg-hls-client shows these for the failed ones:
[http @ 0x5648f6f7e7c0] HTTP error 502 Bad Gateway
http://localhost:8080/vod/avc/WAR_TRAILER_HiQ_10_withAudio-3/index.m3u8: Server returned 5XX Server Error reply

I am running the multistream script via docker as per instructions

measure-quality: accepts YUV - clarify file extensions

https://github.com/intel/media-delivery/blob/master/measure/quality/measure-quality

Need to clarify what we understand under accepting YUV: which file extension, etc.

measure-perf: match bitrates with measure-quality

https://github.com/intel/media-delivery/blob/master/doc/man/measure-perf.asciidoc

Does it make sense to match bitrates w/ measure-quality?

measure/perf: account on empty event lists for Linux perf

Here and below:

media-delivery/measure/performance/MSPerf.py

Line 697 in 50da297

linux_perf_cmdlines = "perf stat -a -e {} -o " + linux_perf_dump + " " + linux_perf_cmdlines

We do not count on the possibility that event lists might be empty which will cause to run measure-perf due to error from Linux perf.

W: GPG error: https://repositories.intel.com/graphics/ubuntu focal InRelease

media-delivery/Dockerfile.ubuntu

Line 36 in dda4d5d

ARG APT_GRAPHICS_REPO="https://repositories.intel.com/graphics/ubuntu focal main"

Step 11/80 : RUN apt-get update && apt-get install --no-install-recommends -y   wget   && rm -rf /var/lib/apt/lists/*
 ---> Running in 77bdf775c000
Get:1 http://security.ubuntu.com/ubuntu focal-security InRelease [107 kB]
Get:2 http://archive.ubuntu.com/ubuntu focal InRelease [265 kB]
Get:3 http://archive.ubuntu.com/ubuntu focal-updates InRelease [111 kB]
Get:4 https://repositories.intel.com/graphics/ubuntu focal InRelease [3195 B]
Get:5 http://security.ubuntu.com/ubuntu focal-security/restricted amd64 Packages [45.4 kB]
Get:6 http://archive.ubuntu.com/ubuntu focal-backports InRelease [98.3 kB]
Get:7 http://security.ubuntu.com/ubuntu focal-security/multiverse amd64 Packages [1078 B]
Ign:4 https://repositories.intel.com/graphics/ubuntu focal InRelease
Get:8 http://security.ubuntu.com/ubuntu focal-security/main amd64 Packages [231 kB]
Get:9 http://archive.ubuntu.com/ubuntu focal/universe amd64 Packages [11.3 MB]
Get:10 https://repositories.intel.com/graphics/ubuntu focal/main amd64 Packages [26.6 kB]
Get:11 http://security.ubuntu.com/ubuntu focal-security/universe amd64 Packages [67.5 kB]
Get:12 http://archive.ubuntu.com/ubuntu focal/multiverse amd64 Packages [177 kB]
Get:13 http://archive.ubuntu.com/ubuntu focal/main amd64 Packages [1275 kB]
Get:14 http://archive.ubuntu.com/ubuntu focal/restricted amd64 Packages [33.4 kB]
Get:15 http://archive.ubuntu.com/ubuntu focal-updates/restricted amd64 Packages [45.7 kB]
Get:16 http://archive.ubuntu.com/ubuntu focal-updates/universe amd64 Packages [207 kB]
Get:17 http://archive.ubuntu.com/ubuntu focal-updates/main amd64 Packages [457 kB]
Get:18 http://archive.ubuntu.com/ubuntu focal-updates/multiverse amd64 Packages [17.3 kB]
Get:19 http://archive.ubuntu.com/ubuntu focal-backports/universe amd64 Packages [3195 B]
Fetched 14.5 MB in 2s (6163 kB/s)
Reading package lists...
W: GPG error: https://repositories.intel.com/graphics/ubuntu focal InRelease: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY 9B10C065DBB72B06

allow all variate of video codecs supported by containers (.mp4)

media-delivery/samples/cdn/do-transcode.sh

Line 132 in 59f235a

-c:v h264_qsv -re -i $to_play -c:a copy

We force video decoder plugin to be h264_qsv which effectively limits the input content we accept for the demo to AVC input. Need to relax this requirement.

Fix function encoder capability mismatched with driver's table.

https://github.com/intel/media-delivery/blob/master/doc/benchmarks/intel-data-center-gpu-flex-series/intel-data-center-gpu-flex-series.rst#fixed-function-encode-vdenc

see https://github.com/intel/media-driver/blob/master/docs/media_features.md#hardware-encoding-low-power-encodingvdenchuc
for hevc 8 bits.

Instead of maintaining a local copy of capability. could we point to driver capability?

thank you

measure/perf: script keeps spawning helper tools

media-delivery/measure/performance/MSGo.py

Line 64 in 9a0df71

os.system("killall top") # Kill Top Application.

killall is missing from docker image setup, so script actually keeps spawning helper tools eating cpu cycles.