Assess learners' ability to identify, measure, and diagnose performance bottlenecks and root causes in cloud-native distributed applications. Scope includes interpreting metrics, logs, and distributed traces; using observability and profiling tools (Prometheus, OpenTelemetry/Jaeger, APMs, profilers, kubectl); analyzing latency percentiles, throughput, resource contention, network and configuration issues, and concurrency; and prioritizing mitigations aligned with SLOs and capacity planning.