The way I wanted to do this is to create dashboards that would serve at the same time as infrastructure diagrams for documentation and live health monitoring.
Right now, most documentation solutions aren't used on a daily basis so become out of date, because people don't think about it when making changes and fixes.
And monitoring solutions only show you charts of things you're supposed to already know. They're very technically-oriented, and not business-logically oriented, if that makes sense. Like they'll tell you that process x is running on machine m, and that it's running out of ram, but nothing will tell you that process y that depends on x's outputs is going to fail as well.