Answer-first summary for fast verification
Answer: Utilizing Azure Databricks' MLflow to train and deploy a machine learning model that predicts anomalies based on historical data
The correct approach involves using MLflow within Azure Databricks to train and deploy a machine learning model for anomaly detection. This method leverages historical data to accurately predict anomalies in real-time, adapting to changing data patterns for more effective detection. Other options, such as static thresholds or built-in statistical functions, lack the sophistication and adaptability of machine learning models. While integrating Azure Stream Analytics with Azure Machine Learning is viable, it introduces unnecessary complexity compared to the streamlined MLflow approach.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
How would you implement advanced anomaly detection and alerting on streaming metrics in a Databricks pipeline monitoring real-time IoT device data?
A
Sending streaming data to Azure Event Hubs and using Azure Stream Analytics with Azure Machine Learning for anomaly detection
B
Implementing Spark Structured Streaming with built-in statistical functions for real-time anomaly detection
C
Configuring static thresholds in Azure Monitor for alerting on key metrics
D
Utilizing Azure Databricks' MLflow to train and deploy a machine learning model that predicts anomalies based on historical data
No comments yet.