Microsoft Azure Data Fundamentals DP-900 Practice Question

Streaming (also called real-time) processing is designed to handle a continuous flow of events with very low latency, so it can evaluate each sensor reading almost immediately and trigger an alert if a threshold is breached. Batch processing-whether hourly, nightly, or weekly-collects data over a period and processes it later, so it cannot satisfy the sub-second or near-real-time responsiveness required for safety alerts. Therefore, streaming is appropriate for the alerting scenario, while batch jobs remain suitable for the separate historical reporting task. Options describing overnight, weekly, or manual ETL jobs are examples of batch workloads and would not meet the company's immediacy goal.