Archive notice: This article was originally published on January 25, 2019. Links and embedded videos are preserved as part of the historical record.
OpenCue is an open-source render-management system that you can use for VFX and animation productions to split larger render jobs into individual tasks. These can be passed into a queue and the required resources automatically assigned. The system is based on Sony Imageworks’ in-house render manager “Cue3“, which they have already used for hundreds of films and have now released as “OpenCue“. With Zync , Google Cloud also brings useful experience in the context of VFX and animation projects to the table. Google and Sony Imageworks are now also members of the Academy Software Foundation, founded at the end of last year.
OpenCue Features
OpenCue offers several functions for managing render jobs, including…
- Scalable architecture for the simultaneous use of numerous machines
- Tagging system to assign certain jobs to machine types
- Jobs are routed to and processed on a central render farm
- Multi-threading support currently for Katana, Prman and Arnold
- Cloud support
- Hosts can be divided into a large number of processes, each with its own reserved core and memory requirements.
Architecture & Use
On the user or artist-workstation side, CueSubmit is used for submitting render jobs and CueGUI for managing running jobs. CueSubmit (as a PySide integration) can be run as a plugin in the respective software – such as Maya, Houdinie or Nuke.
Before the client-side OpenCue tools can be run, PyCue and PyOutline must also be installed. More on that here.
In the background, Cuebot (Job Submission Manager, Distribution, API Requests) and RQD (host registration etc.) run, among other things: You can find installation instructions for the admin here. The current OpenCue source code, executable files and documentation are now available on GitHub under an Appache 2.0 licence. Further tutorials and sample projects will follow in the coming days.
To the Github Repository
Further Information: On the Google Blog
