What I do miss are clear and precise instructions / documentation on how to set up and run a community node. Hardware resources needed, recommended, software needed, set up procedure, day to day operations needed. The n the why 's of running a node - benefits, not only financial.

Perhaps all this is already written and published somewhere?

All this to distribute the load from your cluster to the community.

Next, looking into the future - one word comes to mind - sharded.

If it really is a requirement to have 128 GB of RAM to run a node then that's way too much. Somehow the load should be lowered to come into a reasonable range. I am not telling you anything new, I guess.

OK, I realize that my suggestions are not very constructive. I can't propose an out of the box solution. I can only voice my concern. I believe that you are working on a long term solution to keep the platform functional and scalable.

Good luck!