I've built software which can provide distributed global ordering and mutexes over even geographically distributed kafka topics.
I believe I can bring together the monolith-people with the microservice-people by reconciling the benefits of both approaches.
This is an "easy to access" article on my project:
https://andrasgerlits.medium.com/distributing-consistency-6a1d11edc1cd
This is one on CAP Theorem and how it applies to us
https://andrasgerlits.medium.com/cap-and-us-9e0734cbfe1e
AMA
Does this new database support SQL? I saw a few mentions but not sure if it was explicitly stated
It sounds like the bottleneck is going to be the transaction controller in each node, which has to serialize the processing of transactions to ensure they don't clobber each other. I guess that could still scale pretty well as long as you can scale out the number of nodes?