Google Professional Data Engineer
Get started today
Ultimate access to all questions.
You're overseeing an Apache Kafka-based IoT pipeline that processes 5000 messages per second. Your task is to create an alert that triggers when the one-hour moving average drops below 4000 messages per second. Which Google Cloud Platform strategy should you implement?
You're overseeing an Apache Kafka-based IoT pipeline that processes 5000 messages per second. Your task is to create an alert that triggers when the one-hour moving average drops below 4000 messages per second. Which Google Cloud Platform strategy should you implement?
Explanation:
Option C stands out as the optimal solution for this scenario. By leveraging Dataflow with Kafka IO, you can efficiently process the data stream, applying a sliding one-hour window every five minutes to monitor the moving average. This method promptly alerts you if the average message rate dips below the threshold, ensuring timely detection of rate fluctuations. Alternatives A, B, and D, while feasible, introduce unnecessary complexity or lack the precision of a sliding window approach for moving average calculations.