VM Health Monitoring at 1M Scale

Requirements Track the health of 1,000,000 VMs. Every VM uses one of these mechanisms: It answers a control-plane ping request (pull model), or It sends a heartbeat once every 10 seconds (push model). The monitoring platform must gather and display the status of all 1M VMs during a 5-minute period. The key interview discussion is: identify the limiting factor for each approach and explain why those limits are different. Preparation Practice contrasting the two bottlenecks…