A hospital network carries life-critical monitoring alongside SNMP, syslog, and backup traffic, management sessions drop during congestion, and auditors require reliable device visibility at all times. Which QoS treatment for network management traffic ensures resilient monitoring?

The correct treatment succeeds because SNMP, syslog, and SSH belong to the network-management class, which calls for specific marking coupled to a dedicated queue with a minimum bandwidth guarantee that is protected during congestion, so device visibility and interactive access persist for auditors without taking strict priority away from life-critical monitoring that needs low latency. Placing life-critical traffic in strict priority while leaving management in default best-effort fails by queuing principle because best-effort gets only leftover service with aggressive tail drops, so SSH sessions still collapse and telemetry gaps continue exactly as described. Marking management correctly but forwarding without reserved bandwidth fails by confusing marking with treatment, since DSCP alone only signals intent and a congested scheduler with no minimum guarantee will still drop those packets like any best-effort traffic. Grouping backup together with management in one guaranteed queue fails by aggregation principle because high-throughput bulk TCP backup fills buffer depth and starves small transactional management packets, recreating congestion inside the protected class. For exams, remember network-management gets assured bandwidth but never strict priority, which is reserved for voice and life-critical real-time flows. In practice verify with show policy-map interface during a load test to confirm zero management drops and continuous syslog while backup remains throttled.