Recently I’ve found that there is no simple way to render using cloud (like GCP/Amazon) - everyone actually buying some hardware and building ~$3000-5000 render farms or using some proprietary net-rendering solutions, which are unstable or have none security to be somehow used in a cloud environment.
So I decided to build the “BlendNet” addon for blender 2.8 with GCP provider support - “preemptible” instances could be really useful for the rendering purposes and could offer ~$0.5 per quite complicated cycles frame. GCP is offering $300 for anyone to play with and it’s easy to setup the project and actually run your render without any man-in-the-middle. If the addon will become useful - there is a number of things to add, like cost estimation, another providers to support, etc. So hopefully it will help you guys.
As usual 25% of the plan are taking ~50% of time… But finally I prepared alpha version of the addon - I updated the header with the required information.
So I think it’s quite optimal right now - the same results with the reference progressive render is a huge value of the system. Of course we need to work on support for tile rendering, because 2min difference is pretty big if we will multiply it by thousands of frames. And remaining calculation still requires some improvement…
Hi, that’s the thing - it’s not some addon for proprietary cloud. It just allow you to use the general clouds (Google Cloud for now, and soon Amazon, Azure, Digital Ocean and local farm) without paying the man in the middle (because almost every proprietary render clouds just uses the big cloud providers).
So if you have access to GCP in your country - it will work.
Is it possible to use blendnet to split the render between the available resources on a local machine i.e. send Scene A to the CPU, Scene B to GPU1 and Scene C to GPU2?
Hi @anon71893420, in general about splitting the CPU - I used docker for testing the local provider and mem/cpu limits there is working well.
About GPU - BlendNet is not supporting it officially right now. You can try to select the required render device in settings and use those settings for different Manager/Agents environments… I think it could work, but you defenetely will need to check the script-render.py file, which is actually perform rendering on Agent.
Overall the GPU support is necessary of course, but for now it’s hard to test properly. So if you can try and provide some information - that will be great) Added a new ticket to work on the feature: https://github.com/state-of-the-art/BlendNet/issues/83
Hi, this addon is a great idea, but sadly it is not working for me, I don’t know why.
I’m trying to do it using aws, but once the manager connects and is running, the ‘Run image task’ and ‘run animation tasks’ buttons are disabled and I cannot do anything, even I cannot choose an agent.
Any ideas on how to solve this? I have tried with Blender 2.93.5 and 2.93.4. Am I missing something?
It says it’s connecting, but seems that the manager is already running, and if I check in the aws console there is there blendnet instance running, and I also have a bucket in S3.
Then if I restart Blender it doesn’t say “Manager connecting”, just “idle”, but still the same problem. I can only get the manager running…