AZ-400 Designing and Implementing Microsoft DevOps Solutions Exam

Correlate Usage and Performance Telemetry

What Telemetry Does Application Insights Collect?

Application Insights collects a wide range of telemetry data from both server web apps and client webpages. For server web apps, it gathers:

• HTTP requests: Information about incoming requests to your application.
• Dependencies: Data about external service calls, like SQL databases and Azure Storage.
• Exceptions: Logs of errors occurring in your application.
• Performance counters: Metrics like CPU usage, memory usage, and more, collected via various agents.
• Custom events and metrics: User-defined metrics providing specific insights into business events.
• Trace logs: Detailed logs for deeper investigation of application behaviors.

From client webpages, it collects:

• Uncaught exceptions: Including stack traces and error details.
• Network dependency requests: Information on requests made by the app.
• User, device, and session info: Details about users and their devices.

How to Use Application Insights Effectively

After enabling Application Insights, you can start with Live metrics and the Application map:

• Live metrics: Offer real-time insights into what’s happening with your application, helping you diagnose issues quickly.
• Application map: Visualizes the structure of your application and its dependencies, highlighting bottlenecks and failures.

These features provide a comprehensive view of your application’s performance and can guide you in making necessary adjustments.

Analyzing Metrics for Performance

Monitoring and analyzing performance metrics is key to ensuring your application's reliability and scalability. Here are a few critical areas to focus on:

• Request rates, response times, and failure rates: Identify peak usage times and correlate them with performance anomalies.
• Dependency performance: Find out if external services are causing delays.
• Exceptions and logs: Dive into stack traces and logs to pinpoint issues causing errors.

By understanding these metrics, you can troubleshoot issues before they impact users significantly.

Best Practices for Traces

Enabling trace collection in Application Insights provides a breadcrumb trail of your application’s activities, which is crucial for investigations:

• Integrate Application Insights with your logging framework (e.g., ILogger) to capture trace logs.
• Analyze trace logs alongside exceptions to get contextual information about the errors.
• Ensure each log has contextual identifiers like operation IDs for effective correlation.

These practices help in maintaining a detailed record of application activities and diagnosing problems efficiently.

Setting Up Availability Tests

To ensure your application is always available, use availability tests such as web tests or ping tests:

• Configure these tests in Application Insights to regularly check if your application's endpoints are up and running.
• Specify critical endpoints to monitor in these availability tests.

By regularly collecting and analyzing telemetry data, you can maintain application health, improve performance, and ensure reliability, thereby meeting your service-level objectives.

Conclusion

Collecting and analyzing telemetry data using tools like Azure Monitor and Application Insights is essential for understanding your application's performance and user behavior. Key areas such as monitoring HTTP requests, dependencies, performance counters, and trace logs offer insights into the health of your applications. Using features like Live metrics and Application maps helps visualize issues in real-time. Regularly monitoring request rates, response times, failure rates, and setting up availability tests ensure that your applications remain reliable and perform well under varying conditions. By following these practices, organizations can maintain service-level objectives effectively.