We, as I'm sure a lot of businesses do, have a resilient network. This means that we have multiple core switches in our network, and our edge switches have connections to each of them.
At the moment, dependencies are built on a 1:1 relationship. It would be good if you could create a group to say that this object AND this object (core switch 1 and core switch 2) both have to be down for it to ignore alerts on everything else.
The same thing with WAN sites, many of them have 2 routers. At the moment, if the primary goes down, it stops alerts for everything, even though the site might still be running on the backup/secondary link. (This is the same with many solutions, firewalls, F5s plus probably others).