AZ-400 Designing and Implementing Microsoft DevOps Solutions Exam

Inspect infrastructure performance indicators, including CPU, memory, disk, and network

Infrastructure performance indicators help you understand how your systems are running in real time. By tracking CPU, memory, disk, and network metrics, you can spot bottlenecks before they impact users. Performance indicators are critical for maintaining system health and ensuring applications run smoothly.

Monitoring CPU utilization shows how much processing power your virtual machines or containers are using. High CPU usage over time can indicate that your application needs scaling or optimization. Conversely, very low usage might suggest over-provisioned resources, leading to unnecessary costs.

Memory metrics reveal how much RAM your services consume and help detect memory leaks. When memory use stays near capacity, performance can degrade or instances can crash. Disk metrics, such as read/write operations and latency, tell you if your storage can keep up with demand. Finally, network metrics—like inbound and outbound throughput—help you ensure data transfers are efficient and secure.

To view these indicators in Azure, use Azure Monitor Metrics or dashboards in the Azure portal. You can set up alerts on thresholds for each metric, enabling proactive responses to issues. By regularly reviewing these indicators, you maintain high availability and keep performance costs optimized.

Analyze metrics by using collected telemetry, including usage and application performance

Telemetry consists of data that your applications and services send about their operations. Collected telemetry can include usage statistics and application performance details. This data is essential for understanding user behavior and identifying performance issues that occur during runtime.

Usage metrics, like the number of active users or requests per second, show how customers interact with your application. Application performance metrics—such as response times, error rates, and dependency durations—help you pinpoint where slowdowns occur. When combined, these metrics provide a complete picture of both system and user experiences.

In Azure, you can view these metrics using Azure Monitor or Application Insights. The Metrics Explorer allows you to create charts and dashboards, while workbooks let you layer multiple visuals together. You can also configure alerts based on telemetry, ensuring you are notified when key thresholds are crossed.

Analyzing this telemetry allows you to:

• Detect performance regressions after updates
• Understand feature usage trends
• Identify resource bottlenecks
• Make informed decisions on scaling and optimization

By leveraging telemetry, you turn raw data into actionable insights that drive continuous improvement in your DevOps processes.

Inspect distributed tracing by using Application Insights

Distributed tracing tracks requests as they move through various services and components. Application Insights provides built-in support for this feature, allowing you to see each step of a transaction. This end-to-end visibility helps you find the exact point of failure or slowdown.

In the Application Insights portal, the Transaction Search and Performance blades show trace trees and call stacks. You can click on individual operations to view detailed timings, dependencies, and custom properties. Underline the value of distributed tracing by quickly identifying high-latency operations that hurt user experience.

When you inspect traces, you can correlate client requests with back-end processing and database calls. This correlation allows teams to:

• Isolate slow services
• Understand dependency chains
• Diagnose intermittent errors
• Optimize end-to-end performance

Using distributed tracing with Application Insights ensures your entire application landscape is transparent, making troubleshooting faster and more effective.

Interrogate logs using basic Kusto Query Language queries

Azure Monitor Logs store large volumes of telemetry in a central repository. To explore this data, you use Kusto Query Language (KQL), a powerful yet simple query language. KQL enables you to filter, project, and aggregate log records to answer operational questions.

A basic KQL query starts by specifying a table name, such as AppRequests or Heartbeat, and then adds operations like where, project, and summarize. For example:

• where filters records by conditions
• project selects specific columns
• summarize groups records and calculates counts or averages

You can run these queries in the Azure portal’s Logs blade or integrate them into workbooks and alerts. By mastering basic KQL patterns, you gain flexible log analysis capabilities and the power to craft custom reports.

With KQL, you can:

• Detect error patterns over time
• Correlate logs with metrics
• Create visualizations for dashboards
• Automate incident investigations

Log interrogation using KQL is a key skill for any DevOps engineer looking to maintain and improve application reliability.

Conclusion

In this section, we covered how to inspect infrastructure performance by monitoring CPU, memory, disk, and network metrics to ensure smooth operations. We then discussed how to analyze telemetry for usage and application performance, turning data into actionable insights. Next, we explored distributed tracing through Application Insights, highlighting its importance for pinpointing request bottlenecks across services. Finally, we detailed how to interrogate logs with basic KQL queries, empowering you to filter and visualize log data effectively. Together, these practices form a comprehensive approach to analyzing metrics from instrumentation in Azure, helping you maintain high availability and deliver optimal user experiences.