GCP Associate Cloud Engineer Practice Question 98: Cloud Trace Latency Microservices
Google Cloud Associate Cloud Engineer · Test 10 · question 8 of 10
A request to your application passes through several microservices before a response is returned to the user. Users report that some requests are slow, but the team cannot tell which individual service in the call chain is adding the most latency. They want to visualize the time spent in each service for individual requests.
Which Google Cloud tool should they use?
Show answer and explanation
Answer:
Cloud Trace collects latency data for individual requests as they move across services and displays it as a trace made up of spans, one per service or operation, so engineers can see exactly where time was spent in the call chain for a specific request. That is precisely the distributed latency breakdown the team needs.
Option A is wrong because log-based metrics turn log entries into time series counters or distributions for alerting and dashboards; they do not show the per-request path through multiple services. Option B is wrong because uptime checks only verify that an endpoint is reachable and responding, reporting availability rather than where latency accumulates internally. Option D is wrong because Cloud Profiler analyzes CPU and memory consumption inside a single running application's code, not end-to-end latency across a chain of separate microservices.
Reference: Cloud Trace overview