Follow the procedures below to set up the Debian and the Debian Docker image packages of Pixyz Scenario Processor:

Standalone Debian

Pixyz Scenario Processor is available as a Debian package:

Requirements


  • Ubuntu 20.04 (Focal) or similar
  • It comes as a  tar.gz archive, available for download here


Installation


  • Install the required packages:

    apt-get update
    apt-get install -y libexpat1 libbsd0 libopengl0 libegl1 libglu1-mesa libglx0 --no-remove


  • Upload the tar.gz archive on the instance and extract it to a convenient place
  • Finish the installation process running:

    ./PiXYZFinishInstall PiXYZScenarioProcessor


Standalone Debian Docker Image

Pixyz Scenario Processor is available as a ready-to-use Docker container.


Pulling from Docker Hub


The docker-hub repository is private, access must be granted to pull container:


docker pull pixyzinc/scenarioprocessor:2022.1.0.35


The container uses /usr/bin/PiXYZScenarioProcessor as default entrypoint (ideal for ephemeral use).


Download image archive


Container is available as a compressed archive to download here.


Usage


USAGE: PiXYZScenarioProcessor [options]
Options:
  --help   Print usage and exit.
  --batch  Force batch mode. Otherwise, auto select between batch and scenario
           processor mode.
  --lic    Install license file ( optional ).
Examples:
  Interactive mode :
    PiXYZScenarioProcessor [ --lic=localFile.lic | --batch ]
  Non interactive mode :
    PiXYZScenarioProcessor [ --lic=localFile.lic | --batch ] pythonFile.py
    PiXYZScenarioProcessor [ --lic=localFile.lic | --batch ] pythonFile.py args1
args2 ...
    PiXYZScenarioProcessor [ --lic=localFile.lic ] pluginName functionName
    PiXYZScenarioProcessor [ --lic=localFile.lic ] pluginName functionName args1
args2 ...


License


  • Checkout this page for information about licensing
  • If you are using a license file pixyz.lic to register your instance to the license server, place it in (or mount a Docker volume) /usr/share/PiXYZScenarioProcessor/licenses
  • The license server can also be registered in python scripts at runtime using core.configureLicenseServer("lics-it-gcp-p01", 27005, True)


Plugins


  • If the scenarios you want to run in Pixyz Scenario Processor come as Plugins, place them in (or mount Docker volume) /usr/share/PiXYZScenarioProcessor/plugins


Limitations

Unsupported formats

The following input file formats are not supported with Debian packages of Pixyz Scenario Processor (only Windows):

  • Sketchup format
  • Autodesk Alias
  • Autodesk VRED
  • Autodesk Recap


All supported file formats can be found here


Activating GPU acceleration in Docker image

A few Pixyz algorithms are accelerated on GPU (more information here).

The following link contains instructions to setup nVidia drivers and nVidia-Docker modules for Docker images: https://github.com/NVIDIA/nvidia-docker/wiki/Installation-(Native-GPU-Support)#prerequisites
Once the machine is setup, you will simply need to add the "--gpus all" option in your docker run command (docker run --gpus all ...)