If you want to test the ERM policies in a controlled environment, it's almost mandatory to have tools that allow you to overload the router. One of these tools is the UDP flood: if you flood a router's IP address, you're guaranteed to raise the CPU to 100%, with majority of the process CPU being used by the IP Input process (the interrupt CPU load will also be significant).
This phenomenon illustrates very clearly why it's so important to have inbound access lists protecting the router's own IP addresses on all edge interfaces.
If you want to stress-test the router's forwarding functionality, you could use the host route to the null0 interface; all packets sent to that IP address will be CEF-switched, so the only impact of the UDP flood to the unreachable IP address will be the increased interrupt CPU load. I was able to increase the interrupt CPU load to close to 50% on a 2800 router using a virtual PC with a Fast Ethernet interface.
And just in case you need it, here is the configuration from my test router. All packets sent to 10.0.0.22 are CEF-switched and dropped (the CPU load from the IP input process is negligible).
interface FastEthernet 0/0
ip address 10.0.0.1 255.255.255.0
ip route 10.0.0.22 255.255.255.255 null 0