AZ-204 Practice Question: Application Insights adaptive sampling to reduce…
A high-traffic API sends approximately 80,000 telemetry events per second to Application Insights. Monthly ingestion costs are too high. The team wants to reduce data volume by roughly 80 percent while still seeing representative samples of all request types. What should the developer configure?
⚠ Common exam trap
Candidates often confuse ingestion sampling (a portal-level fixed filter) with adaptive sampling (an SDK-level dynamic filter), assuming any sampling in the portal will suffice, but only adaptive sampling can meet the dual goals of volume reduction and representativeness across all request types.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Enable adaptive sampling in the Application Insights SDK and set a target events-per-second limit
Adaptive sampling in the Application Insights SDK automatically adjusts the volume of telemetry sent to the service, targeting a specified rate of events per second. By setting a target that reduces the original 80,000 events/sec to roughly 20%, the SDK will intelligently sample all request types proportionally, preserving representative data while cutting costs.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Enable adaptive sampling in the Application Insights SDK and set a target events-per-second limit
Why this is correct
Adaptive sampling runs in the SDK. It monitors the outgoing telemetry rate and automatically raises or lowers the sample percentage to stay near the target rate. All operation types are sampled proportionally, so statistical trends remain meaningful even at 20 percent of raw volume. Data is reduced before transmission, lowering both network and ingestion costs.
- ✗
Configure ingestion sampling in the Azure portal to retain 20 percent of incoming telemetry
Why it's wrong here
Ingestion sampling discards 80 percent of events after they arrive at the Application Insights endpoint. The data is still transmitted from the application, so network bandwidth is unchanged. It reduces storage and query costs but not the cost of transmitting data from the app.
- ✗
Increase the TelemetryClient flush interval from 30 seconds to 5 minutes to batch events
Why it's wrong here
Flush interval controls when buffered events are sent, not how many are sent. Flushing less frequently does not reduce the number of events — it only changes the timing. Total event volume and ingestion cost remain the same.
- ✗
Filter all events with HTTP status code 200 from the telemetry pipeline before sending
Why it's wrong here
Filtering out 200-OK requests removes all evidence of successful operations. Metrics like response time percentiles, throughput, and dependency success rates become meaningless. Sampling preserves proportional representation; hard filtering destroys observability for the majority of traffic.
Go deeper
Related to this question
About these practice questions
This AZ-204 question is part of Courseiva's 881-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AZ-204 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AZ-204 exam.