Design a High CPU Incident Response System

Design a system for diagnosing and resolving high CPU incidents in a distributed service cluster, including tools for identifying affected hosts, analyzing traffic patterns, detecting retry storms, and managing recovery actions like restarting or scaling nodes.