Optimize and Troubleshoot Data Storage and Data Processing DP-203 Practice Questions (Page 4)

Questions 16–20

1. 16application · medium

   A Spark streaming job in Azure Databricks writes aggregated counts by user ID to a Delta table. The job's micro-batches are slow, and the streaming UI shows that a few tasks process far more data than others. You need to balance the workload across executors. What should you do?

   • AEnable speculative execution in the streaming job.
   • BAdd a random salt to the user ID and perform a two-stage aggregation in the streaming query.
   • CIncrease the number of streaming micro-batch intervals.
   • DUse a stateful streaming aggregation with a longer watermark.
   Reveal Solution Discussion Select an answer first
2. 17foundation · easy

   What is the primary benefit of using workload isolation in a dedicated SQL pool?

   • AIt automatically compacts small files in the data lake.
   • BIt replicates data to a secondary region for disaster recovery.
   • CIt prevents one workload from consuming all resources and impacting other workloads.
   • DIt encrypts data in transit between the client and the server.
   Reveal Solution Discussion Select an answer first
4. 18foundation · easy

   What is the purpose of result-set caching in Azure Synapse Analytics?

   • ATo store the query text for auditing purposes.
   • BTo store query results so that repeated queries can return data faster without recomputation.
   • CTo replicate data to a secondary region for disaster recovery.
   • DTo automatically tune indexes based on query patterns.
   Reveal Solution Discussion Select an answer first
5. 19application · medium

   A Power BI dashboard queries an Azure Synapse Analytics dedicated SQL pool. The dashboard is accessed by many users, and the underlying data changes only once a day. Users report slow load times. You need to reduce query latency for the dashboard. What should you do?

   • ACreate a nonclustered index on every column used in the dashboard's filters.
   • BUse Azure Synapse Link to replicate the data to a dedicated SQL pool in real time.
   • CEnable result-set caching on the dedicated SQL pool.
   • DIncrease the service level of the dedicated SQL pool to DWU2000c.
   Reveal Solution Discussion Select an answer first
6. 20foundation · easy

   A data engineer notices that a Spark job reading from a directory with thousands of small files runs slower than expected. Which technique directly addresses the performance impact of many small files?

   • ACompact the small files into fewer, larger files before processing.
   • BIncrease the number of Spark executors to process more files in parallel.
   • CSwitch from Parquet to CSV file format for the source data.
   • DUse a larger file block size in the underlying storage system.
   Reveal Solution Discussion Select an answer first