Where is interesting operating system research happening these days?

There are a lot of projects that are looking at blurring the lines for systems that run databases:

https://dbos-project.github.io/blog/intro-blog.html https://www.cs.cit.tum.de/fileadmin/w00cfj/dis/papers/cumulu... .

Plenty of scheduling interest: https://research.google/pubs/ghost-fast-and-flexible-user-sp...

We ourselves implemented UMCG for nanos: https://nanovms.com/dev/tutorials/user-mode-concurrency-grou... .

The formally verified seL4 kernel: https://sel4.systems/About

Plan9 broke some new ground as well.

What holds back any progress with OS research is the requirement to run existing software. For example, Haiku-OS (which is a re-engineering of BeOS) is failing to gain much traction because people want to run Firefox, LibreOffice, etc on it. And of course, with a novel OS architecture, it requires an adaption layer. Which in turn largely negates the impact of any new OS architecture.