Databricks-ML-Pro ML Ops Practice Question
A fraud detection model was deployed to a Databricks Model Serving endpoint. Over the past week, the endpoint's p99 latency has increased from 45 ms to 320 ms, but the model's predictions remain accurate. The serving logs show that each request now includes a larger JSON payload with additional transaction metadata. What is the most likely cause of the degradation?
⚠ Common exam trap
The trap here is assuming that latency degradation always indicates infrastructure scaling problems, ignoring that request payload changes can increase per-request processing time.
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 model's feature vector has grown, and the scoring function is performing expensive feature transformations on the raw payload before inference.
The increased payload size likely triggers additional feature engineering or parsing inside the model's scoring logic, raising per-request CPU time. Since accuracy is unchanged, the model itself is fine; the bottleneck is the preprocessing step. The other explanations either contradict the payload-size correlation or would impact accuracy or availability.
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 serving endpoint is hitting the maximum number of concurrent requests and queuing requests, which adds wait time.
Why it's wrong here
Queuing due to concurrency limits would affect latency under load, but the scenario states that each request now includes a larger payload, pointing to per-request processing time rather than queue wait. Concurrency issues would also likely cause timeouts or errors, not just a latency increase while accuracy remains stable.
- ✓
The model's feature vector has grown, and the scoring function is performing expensive feature transformations on the raw payload before inference.
Why this is correct
When the request payload includes extra fields, the model's predict function or preprocessing logic may compute additional features, increasing CPU time per request. This directly explains the latency increase while accuracy stays stable. The other options either would affect accuracy or are not supported by the symptom of larger payloads.
- ✗
The model artifact is being re-downloaded from the MLflow Model Registry on every request due to a misconfigured artifact cache.
Why it's wrong here
Model Serving loads the model artifact once at endpoint startup and caches it; it does not re-download per request. A misconfigured cache would typically cause startup failures or consistent high latency from the beginning, not a gradual increase after payload size changed. This option is plausible but does not match the observed pattern.
- ✗
The model's underlying compute cluster has been scaled down, reducing available CPU resources for inference.
Why it's wrong here
Model Serving endpoints scale automatically based on load, and a scale-down would affect all requests uniformly. The scenario ties latency growth to larger payloads, not to resource reduction. While scaling issues can cause latency, the specific correlation with payload size makes this less likely than increased preprocessing work.
About these practice questions
This Databricks-ML-Pro question is part of Courseiva's 300-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 Databricks exam blueprint
This Databricks-ML-Pro practice question is part of Courseiva's free Databricks 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 Databricks-ML-Pro exam.