1Z0-1067-25 Question 121
Single answerYour DevOps team is investigating CPU usage spikes across all compute instances in the 'Production' compartment. You must create a metric query using MQL that calculates the average CPU utilization for all these compute instances over the past hour. Which MQL expression meets these requirements?
- A
CPUUtilization[1h]{compartmentName = 'Production'}.mean()
- B
CPUUtilization[1d]{compartmentName = 'Production'}.sum()
- C
CPUUtilization[1h]{compartmentName = 'Production'}.rate()
- D
MemoryUsage[1h]{compartmentName = 'Production'}.mean()
Show answer and explanation
Correct answer: A
Explanation
When creating MQL expressions in Oracle Cloud Infrastructure, you specify the metric name, time window, and dimension filters, followed by the appropriate aggregator. In this scenario, CPUUtilization[1h]{compartmentName = 'Production'}.mean() is the best choice to calculate the average CPU usage for the last hour. For more details, see the Oracle documentation on using the Metric Query Language (MQL): https://docs.oracle.com/en-us/iaas/Content/Monitoring/Concepts/metricquerylanguage.htm
- A. Correct.
Option 1 is correct because it uses the CPUUtilization metric over a 1-hour window, applies the filter for the 'Production' compartment, and uses mean(), which calculates the average CPU usage across all matching resources.
- B. Incorrect.
Option 2 is incorrect because it uses a 1-day time window (1d) and sum() aggregator, which would not provide the average CPU usage over the past hour. sum() would give the cumulative metric rather than an average.
- C. Incorrect.
Option 3 is incorrect because rate() calculates the rate of change rather than the average usage. This is useful for metrics where you want to see the difference over time, but it does not provide an average CPU utilization.
- D. Incorrect.
Option 4 is incorrect because it references the MemoryUsage metric instead of CPUUtilization, so it does not address the CPU usage spikes.