hardMultiple Choice
350-401 Practice Question: Is designing a model-driven telemetry solution…
A network engineer is designing a model-driven telemetry solution for a large enterprise network with thousands of devices. The engineer wants to minimize the load on the network devices and the collector by sending data only when significant changes occur. The engineer decides to use on-change subscriptions. However, after deployment, the engineer notices that some subscriptions are sending updates too frequently, causing high CPU usage on the devices. What is the most likely reason for this excessive update frequency?
⚠ Common exam trap
The trap here is assuming that on-change subscriptions are always low-overhead; candidates forget that the update rate depends entirely on how volatile the subscribed YANG leafs are, not on the subscription mode itself.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
The YANG paths include high-frequency changing leafs like interface counters or CPU load
On-change subscriptions in model-driven telemetry (MDT) trigger an update whenever a subscribed YANG leaf changes value. If the subscription path includes volatile leafs such as interface counters, CPU load, or queue statistics that change on every packet or polling cycle, the device will emit updates almost continuously, driving up CPU and collector load. The fix is to subscribe to stable configuration or state leafs, or use periodic subscriptions with a sample-interval for high-churn data.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
The engineer configured a sample-interval in addition to on-change, causing both periodic and on-change updates
Why it's wrong here
A sample-interval alongside on-change does not cause excessive on-change updates; the device still emits only on genuine change, so the periodic stream is separate. It tempts because mixing subscription types is a plausible misconfiguration, and would be correct if the symptom were steady periodic telemetry rather than bursts.
- ✓
The YANG paths include high-frequency changing leafs like interface counters or CPU load
Why this is correct
On-change subscriptions trigger updates whenever a subscribed leaf changes. Selecting YANG paths containing rapidly fluctuating leafs such as interface counters or CPU load causes near-continuous notifications, driving excessive device CPU usage — precisely the symptom described in the stem.
- ✗
The collector is overwhelmed and sending back-pressure signals causing retransmissions
Why it's wrong here
Back-pressure from the collector cannot make devices generate extra on-change notifications; the collector's state does not drive the device's change-detection logic. It tempts because collector overload is a real telemetry scaling concern, and would be the answer if the symptom were dropped or delayed updates rather than excessive device CPU.
- ✗
The engineer used JSON encoding instead of GPB, causing larger payloads and more CPU usage
Why it's wrong here
JSON encoding increases payload size and collector-side parsing cost, but does not raise the number of on-change notifications or device CPU from change detection. It tempts because encoding choice genuinely affects telemetry efficiency, and would be the answer if the symptom were bandwidth or collector processing overhead.
Go deeper
Related to this question
Learn chapter
Spine-Leaf and Software-Defined Network Architectures
Key term
NETCONF Protocol
NETCONF is a network management protocol that uses a structured data format to configure, retrieve, and modify network devices in a standard, programmatic way.
Key term
REST API for Network Devices
A REST API for network devices is a set of rules that allows software applications to communicate with routers, switches, and firewalls using standard web methods like GET, POST, PUT, and DELETE over HTTP or HTTPS.
About these practice questions
This 350-401 question is part of Courseiva's 1,923-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Cisco exam blueprint
This 350-401 practice question is part of Courseiva's free Cisco certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the 350-401 exam.