Your question is Monitoring Strategy for High Availability. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
How would you design a monitoring strategy for a distributed application to ensure high availability for ICE-scale services?