GCP Associate Cloud Engineer Practice Question

Your e-commerce platform runs on a Cloud SQL for PostgreSQL instance. This morning CPU utilization spiked and customers report slow checkouts. You open Query Insights and see thousands of statements executed in the last hour. The nightly reporting job connects using the database user "reporting_user." What is the fastest way to isolate the queries generated by that job in Query Insights so you can analyze their performance?

Filter the view by Client IP to the VM that runs the reporting job and review the resulting Top queries table.
Switch the display to execution plans and manually expand each plan node to find statements with the largest total CPU time.
Select the User dimension in Query Insights, choose reporting_user, and examine the filtered Top queries list for high-latency statements.
Enable the Cloud SQL Insights API, wait for the next collection cycle, export pg_stat_statements data, and search it offline for reporting_user queries.

Report Issue

Answer Description

Query Insights automatically tags every statement with several dimensions, including the database user that executed it. By selecting the User dimension and filtering for the specific account (reporting_user), the interface immediately narrows both the time-series charts and the Top queries table to only those statements executed by that user. This provides an instant, focused view of the workload created by the reporting job, allowing you to identify which of its queries are consuming the most CPU or exhibiting high latency. Filtering on client IP, downloading raw statistics, or creating external log-based metrics would all take more time and effort and still might mix in unrelated queries. Drilling into execution plans without first narrowing the query set would also be inefficient when thousands of statements are present.

Ask Bash

Bash is our AI bot, trained to help you pass your exam. AI Generated Content may display inaccurate information, always double-check anything important.