CERTIFIED-MACHINE-LEARNING-PROFESSIONAL · Question #21
CERTIFIED-MACHINE-LEARNING-PROFESSIONAL Question #21: Real Exam Question with Answer & Explanation
The correct answer is A: Batch serving has built-in capabilities in Databricks Machine Learning. There appears to be an issue with the provided answer key - E is the correct answer, not A. Why E is correct: When feature values are available a week before query time, predictions can be pre-computed and stored in a database. At query time, the system simply looks up the stored
Question
A machine learning engineer needs to deliver predictions of a machine learning model in real- time. However, the feature values needed for computing the predictions are available one week before the query time. Which of the following is a benefit of using a batch serving deployment in this scenario rather than a real-time serving deployment where predictions are computed at query time?
Options
- ABatch serving has built-in capabilities in Databricks Machine Learning
- BThere is no advantage to using batch serving deployments over real-time serving deployments
- CComputing predictions in real-time provides more up-to-date results
- DTesting is not possible in real-time serving deployments
- EQuerying stored predictions can be faster than computing predictions in real-time
Explanation
There appears to be an issue with the provided answer key - E is the correct answer, not A.
Why E is correct: When feature values are available a week before query time, predictions can be pre-computed and stored in a database. At query time, the system simply looks up the stored result rather than running the model on-the-fly, which is significantly faster. This is the fundamental value proposition of batch serving: trading freshness for speed and efficiency.
Why the other options are wrong:
- A - "Built-in capabilities in Databricks" is a platform marketing claim, not a principled reason to choose batch over real-time serving. Even if true, it doesn't explain why batch serving benefits this specific scenario.
- B - Incorrect; batch serving has clear advantages when features are known well in advance and latency matters.
- C - This is actually an argument against batch serving - real-time predictions would be more current, not batch.
- D - False; real-time serving deployments can absolutely be tested using staging endpoints, shadow mode, or canary deployments.
Memory tip: Think of batch serving as a "meal prep" strategy - you cook everything Sunday (batch), so meals are instant on weeknights (fast lookup). Real-time serving is like cooking to order: fresher, but slower. When ingredients (features) are ready a week early, meal-prepping wins on speed.
Note to exam taker: If this came from a practice test, double-check the answer key - the marked answer (A) appears to be an error. The scenario precisely describes the classic batch serving speed advantage, which maps directly to E.
Community Discussion
No community discussion yet for this question.