Michael G

So, basically the title. What we read in the docs and what our designer said, is that "we can buy one floating Houdini Engine license and render our workflow (split into batches) using any number of rendering nodes". That is "on a farm", which may be the thing that I misunderstood. So we took a Houdini Engine Studios license, set up the licensing server and made it available for our rendering machines (gosh, that was a pain, SideFX guys, your docs are the awful place for starter users), and we want to run hython "optimizer.py", which is basically houRenderNode.render(). This (expectedly) takes a license from a server, runs hython, BUT stops all other containers (lol, SideFX even published a docker container for that, thanks!) from getting this same license, because it is already in use by container #1. I saw some notes on HBatch making some workflow files (IFDs if I am not mistaken?) and distributing them between nodes or something like that, but I am unsure if it would work and if it is able to render what we want (our render node outputs .obj and .png textures, not rendered images of scenes, which I believe Mantra does) (what Mantra is at all? So misleading for a developer, far from design topics, sorry).
So, is the workflow that takes a bunch of files, splits them between containers, and processes them simultaneously possible on a Houdini Engine license, or do I really need 800$ license for each container (I have 40 of those, dynamically scaling up, based on a dataset)? If I can do it using HBatch or any other H<NAMEOFTHETOOL>, can you please point me to the guide on how to set this up, so my containers can render things without waiting for each other to complete. They also do a lot of other things, and being blocked is not what I want to get. Thanks in advance!