Design and Implement Loading Patterns DP-700 Practice Questions (Page 2)

Questions 6–10

1. 6application · medium

   A media streaming company wants to ingest clickstream events from their website into Fabric for real-time personalization. The events are sent to an event hub. The solution must handle high throughput and allow the data to be queried by downstream analytics. What should you implement?

   • AUse a Dataflow Gen2 to read from the event hub and write to a SQL table.
   • BUse a scheduled Pipeline to copy events from the event hub to a Lakehouse every 5 minutes.
   • CUse a Spark structured streaming job to read from the event hub and write the raw events to a Delta table in a Lakehouse.
   • DUse a Power BI direct query to connect to the event hub and visualize the data live.
   Reveal Solution Discussion Select an answer first
2. 7application · medium

   A social media analytics company wants to ingest posts from a public API into Fabric for sentiment analysis. The API provides a continuous stream of posts. The team needs to store the raw posts and also compute sentiment scores in near real-time. Which pattern should the engineer implement?

   • AUse a Dataflow Gen2 to copy the API data to a lakehouse every hour.
   • BUse a scheduled notebook to pull new posts every 5 minutes and process them in batches.
   • CUse an Eventstream to ingest the API data, write raw posts to a lakehouse, and use a Spark streaming job for sentiment analysis.
   • DUse a KQL database to ingest the stream and run sentiment analysis in KQL.
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A manufacturing company needs to load a 500 GB CSV file from Azure Data Lake Storage into a Fabric Lakehouse table every month. The file is a full snapshot of product inventory. The load must complete within the nightly maintenance window of 2 hours. What should you do to load the data most efficiently?

   • AUse a Spark notebook to read the CSV with inferSchema enabled and write it to the Lakehouse table in Delta format using optimized write and compaction.
   • BUse a Dataflow Gen2 to read the CSV and write it to the Lakehouse table with default settings.
   • CUse a Pipeline with a Copy activity to copy the CSV directly into the Lakehouse table without any transformation.
   • DLoad the CSV into a SQL endpoint using BULK INSERT and then create a Lakehouse table from the SQL table.
   Reveal Solution Discussion Select an answer first
5. 9foundation · easy

   Which transformation is commonly applied to data before loading it into a dimension table?

   • AAggregating sales amounts by month
   • BRemoving all foreign key columns
   • CConverting fact rows into key-value pairs
   • DDeduplicating rows to ensure each dimension member appears once
   Reveal Solution Discussion Select an answer first
6. 10expert · hard

   A utility company needs to ingest smart meter readings from millions of devices into Fabric. The readings are sent to an event hub every 5 minutes. The solution must support real-time anomaly detection and also store the data for monthly billing. The team has a limited budget and wants to minimize compute costs. What should they implement?

   • AUse a scheduled batch job every 5 minutes to read from the event hub and process the readings.
   • BUse a Spark structured streaming job with a 1-minute trigger to process readings and write to a Delta table, and use a separate streaming job for anomaly detection.
   • CUse a Fabric eventstream to ingest the readings and write them to a Lakehouse table, and use a separate scheduled job for anomaly detection.
   • DUse a Spark structured streaming job that runs continuously to process all readings and write them to a Delta table.
   Reveal Solution Discussion Select an answer first