Microsoft Fabric Data Engineer Associate DP-700 Practice Question

A KQL database that uses streaming ingestion can accept millions of events per second with minimal configuration, stores the data in native columnar format for efficient long-term retention, and makes new events queryable within a few seconds. Once ingested, the Kusto engine can return interactive KQL queries in sub-second time, satisfying the operations team's analytics requirement. While Spark Structured Streaming can process data with very low end-to-end latency, it generally requires additional development to persist high-volume time-series data in a structure that KQL can query efficiently. Dataflows Gen2 and pipeline copy activities are oriented toward batch or micro-batch workloads and do not meet the throughput and immediate KQL querying needs.