The alert status of one monitor can be made dependent on the alert status of any node that is a member of the same gateway.
Imagine monitoring a router for a single network. If the router goes down the monitor you've set up to test that router will correctly change, first to a Failed state, then to an Alarm state. Unfortunately all the other assets on that same network depend on that same router. When the router fails to connect, those dependent assets can't help but fail to connect as well. An entire branch of the monitor tree reports monitoring failures even though the problem is really a single asset. Those dependent assets are just a distraction at this point. Using dependency relationships you can prevent Network Monitor from triggering a cascade of unnecessary Alarm states when the Alarm state for a single critical monitor will serve the same purpose.
Another example is making all monitors on a single asset dependent on the Ping check monitor. If the network connection to the asset fails, then only one alarm will be created for the Ping check, but not for all the other monitors assigned to that asset.
Click Edit for any gateway, group or asset node, then click the Advanced tab. Use Asset dependency settings to select the monitor this node should be dependent on. All descendants of this node set to inherit will be dependent on the same monitor you select.