Triplox
I have been working on a Datalog engine on top of object storage. The system is called https://triplox.xyz/. I am using SlateDB at the storage layer. The system is very much inspired by Datomic. in its final version the only source of the truth and primitive used should be object storage. The most interesting and experimental part of Triplox, are incremental queries (https://triplox.xyz/incremental-queries/overview/). You get the full power of Datalog with incremental query support. Views or other kind of streaming solutions can be built on top of them. I have also been experimenting with Worst-case optimal joins for the standard engine.