Metrics for resource performance in Azure - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ service_behavior

intermediate

Understanding Azure Monitor Metrics Collection

Which of the following statements correctly describes how Azure Monitor collects metrics for a virtual machine?

A Azure Monitor only collects metrics when the VM is restarted.

B Azure Monitor collects metrics by installing an agent on the VM that sends data every 5 minutes by default.

C Azure Monitor collects metrics directly from the Azure fabric without any agent, providing near real-time data.

D Azure Monitor requires manual export of logs from the VM to collect metrics.

💡 Hint

Think about how Azure Monitor integrates with Azure resources without extra setup.

Azure Monitor collects metrics directly from the Azure platform for most resources, including VMs, without needing an agent. This allows near real-time monitoring.

❓ Architecture

intermediate

Designing Metric Alerts for Azure Resources

You want to create an alert that triggers when CPU usage on an Azure App Service exceeds 80% for 10 minutes. Which metric and configuration should you use?

A Use the 'CPU Percentage' metric with a threshold of 80%, aggregation type 'Average', and evaluation period of 10 minutes.

B Use the 'CPU Percentage' metric with a threshold of 80%, aggregation type 'Maximum', and evaluation period of 1 minute.

C Use the 'CPU Time' metric with a threshold of 80%, aggregation type 'Total', and evaluation period of 5 minutes.

D Use the 'CPU Usage' metric with a threshold of 80%, aggregation type 'Minimum', and evaluation period of 10 minutes.

💡 Hint

Consider the metric name and aggregation that best reflects sustained high CPU usage.

The 'CPU Percentage' metric with 'Average' aggregation over 10 minutes accurately detects sustained CPU usage above 80%.

❓ security

advanced

Securing Metric Data Access in Azure Monitor

Which Azure feature allows you to restrict access to metric data collected by Azure Monitor to only authorized users?

A Azure Role-Based Access Control (RBAC) with roles scoped to the resource or resource group.

B Azure Firewall rules applied to the Azure Monitor service.

C Network Security Groups (NSGs) blocking metric data endpoints.

D Azure Policy enforcing encryption of metric data.

💡 Hint

Think about how Azure controls user permissions for resource data.

Azure RBAC controls who can view or manage metric data by assigning roles at resource or resource group levels.

✅ Best Practice

advanced

Optimizing Metric Retention and Cost in Azure Monitor

You want to retain detailed metrics for 30 days but reduce storage costs for older data. What is the best approach?

A Set metric retention to 7 days and export all metrics to a Log Analytics workspace indefinitely.

B Configure Azure Monitor to retain detailed metrics for 30 days, then archive older metrics to a storage account.

C Disable metric collection after 30 days to save costs.

D Use Azure Monitor's default retention and do not export metrics.

💡 Hint

Consider balancing detailed monitoring needs with cost-effective long-term storage.

Retaining detailed metrics for 30 days in Azure Monitor and archiving older data to cheaper storage optimizes cost and data availability.

🧠 Conceptual

expert

Interpreting Metric Aggregation Behavior in Azure Monitor

You observe that the 'Disk Read Bytes' metric for a VM shows a sudden spike when viewed with 'Total' aggregation over 1 minute, but a smooth curve with 'Average' aggregation. Why does this happen?

A Azure Monitor caches metrics and delays 'Average' aggregation, causing smoother curves.

B The 'Average' aggregation only shows the highest value, hiding spikes; 'Total' shows the sum which is always lower.

C The metric is incorrectly configured; spikes indicate data errors in 'Total' aggregation only.

D The 'Total' aggregation sums all read bytes in the minute, causing spikes when bursts occur; 'Average' smooths these by averaging over the period.

💡 Hint

Think about how summing versus averaging data points affects the graph shape.

'Total' aggregation sums all values in the time window, so bursts cause spikes. 'Average' smooths values by calculating the mean, showing a smoother trend.