BenchMark Quizzes by MDFT Pro

Reduce Telemetry Volume in Application Insights

You work as a Performance Engineer for MDFT Pro, a well-known training agency that provides online certification courses to thousands of students globally. Mark, the Platform Engineering Manager, has deployed a Course Management application built with Java to Azure, and the application has been instrumented using the Application Insights SDK to monitor performance and usage.

The application serves students across multiple regions, generating significant telemetry data from page views, API calls, database queries, and custom events tracking student progress and quiz submissions. The telemetry volume has become expensive, and Mark wants to reduce Application Insights costs by sending less data. However, it’s critical that the monitoring dashboards and performance statistics remain accurate for tracking student engagement metrics and identifying performance bottlenecks.

Which Application Insights SDK feature should you use to reduce telemetry volume without affecting the accuracy of your statistics?

Choose the correct answer from the options below.

Explanations for each answer:

Sampling is correct. Sampling reduces the volume of telemetry sent to Application Insights while preserving statistically correct analysis. It intelligently keeps related telemetry items together, allowing you to navigate between them when diagnosing problems. The sampling algorithm ensures that statistics like request rates and exception rates remain accurate despite sending less data.
Telemetry initializer is incorrect. Telemetry initializers enrich telemetry by adding or modifying properties before it's sent to Application Insights. They don't reduce telemetry volume; instead, they add more data to each telemetry item, which would actually increase the total data sent.
Telemetry processor is incorrect. Telemetry processors can filter out specific telemetry items completely, which does reduce volume, but this affects statistics. When you filter out certain items, you lose data and your metrics become incomplete. This approach doesn't maintain statistical accuracy like sampling does.
Telemetry channel is incorrect. Telemetry channels are responsible for buffering and sending telemetry data to Application Insights. They manage how data is transmitted but don't reduce the volume of data being sent or affect statistical accuracy.

Learn more about Telemetry Sampling:

Application Insights Sampling

Practice Exam

Reduce Telemetry Volume in Application Insights

Discuss this question on social media: