When setting threshold levels for a monitored component, it would be very useful to be able to set the warning and critical levels dependent on another component(s) status. For example, on my SQL Server, say Forwarded Records/Sec is reading 20. In isolation, that means nothing. If I'm doing 400 Batch Requests/Sec it's fine, but if Batch Requests/Sec is 50, then that is a big problem. It would be much better to be able to add some basic relational logic to these fields, either by referencing another already monitored component, or use of a query. This allows for a much better at-a-glace reporting of Green/Yellow/Red dots.
Here's a mock-up of what I am talking about: (Top portion is how it is now, bottom is the proposed)
