Google PCA Ensure solution and operations reliability Practice Question
A company monitors their application with Cloud Monitoring. They set up an alerting policy to notify the on-call team when the 99th percentile latency exceeds 500 ms for 5 minutes. However, they receive false positive alerts due to short bursts. How should they refine the policy?
⚠ Common exam trap
Google Cloud often tests the misconception that lowering thresholds or changing percentiles reduces false positives, when in reality the evaluation window duration is the key lever for filtering out short-lived bursts without sacrificing sensitivity to sustained issues.
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
✓
Increase the evaluation window to 10 minutes.
Increasing the evaluation window to 10 minutes smooths out short bursts of high latency, ensuring the alert triggers only when the 99th percentile latency exceeds 500 ms for a sustained period. Cloud Monitoring evaluates metrics over the specified window, so a longer window reduces false positives from transient spikes while still detecting genuine degradation.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Set up alerting on each data point individually.
Why it's wrong here
Alerting on each data point individually removes the five-minute aggregation window, so every momentary spike triggers a notification and burst-induced false positives multiply. Per-point alerting is right for metrics where any single sample genuinely signals failure, such as a binary health check.
- ✗
Decrease the threshold to 400 ms.
Why it's wrong here
Lowering the threshold to 400 ms makes the policy fire on smaller latency deviations, increasing the false positives caused by short bursts rather than suppressing them. A lower threshold is genuinely useful when you want earlier warning of sustained degradation, not when bursts must be filtered out.
- ✗
Change the metric to average latency instead of 99th percentile.
Why it's wrong here
Average latency smooths away the tail behaviour the 99th percentile was chosen to expose, so genuine slow requests for real users go unnoticed. Average latency suits capacity and trend dashboards, but it cannot replace percentile-based alerting for user-facing latency SLOs.
- ✓
Increase the evaluation window to 10 minutes.
Why this is correct
Extending the evaluation window to 10 minutes requires latency to breach 500 ms across a longer sustained period, filtering out brief spikes that triggered false positives. This directly addresses the stem's short-burst problem while retaining detection of genuine sustained degradation.
Go deeper
Related to this question
Learn chapter
Resource Monitoring and Logging with Cloud Operations
Key term
Latency
Latency is the time delay between a request being sent over a network and the response being received, often measured in milliseconds.
Key term
Cloud Monitoring
Cloud monitoring is the process of observing, measuring, and managing an organization's cloud infrastructure and applications to ensure performance, availability, and security.
About these practice questions
Courseiva writes every PCA question from scratch — 807 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PCA practice question is part of Courseiva's free Google Cloud 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 PCA exam.