Yes, I think the initial simulation definition/set up can be "easy", but once you get to parallel shared memory clusters it unfortunately have to be much more h

Yes, I think the initial simulation definition/set up can be "easy", but once you get to parallel shared memory clusters it unfortunately have to be much more hands on. I had hoped to eventually not only allow running individual solvers from a single interface, but also allow coupling them similar to PreCICE [1], however making that easy to use is something I've not dared attempt yet. [1]: https://www.precice.org https://www.precice.org