Implement a Partition Strategy DP-203 Practice Questions (Page 6)

Questions 26–30

1. 26expert · hard

   A media company stores video viewing logs in Azure Data Lake Storage Gen2. The data is used for two purposes: (1) real-time dashboards that filter by video_id and by view_date, and (2) monthly reports that aggregate all data. The current layout is /logs/2024/05/15/logs.parquet. Real-time queries are slow. You need to implement a partition strategy that improves real-time query performance without significantly increasing storage costs or maintenance overhead. What should you do?

   • APartition by video_id and view_date, e.g., /logs/video_id=123/date=2024-05-15/logs.parquet.
   • BPartition only by view_date, e.g., /logs/date=2024-05-15/logs.parquet.
   • CConvert the files to CSV format for better compression.
   • DKeep the current layout and rely on Azure Synapse Serverless SQL's automatic caching.
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   What is a common partition strategy for streaming data that is being written to a data lake for later batch analysis?

   • APartition by random hash of the event payload
   • BPartition by source IP address
   • CPartition by event size
   • DPartition by event timestamp (e.g., year/month/day/hour)
   Reveal Solution Discussion Select an answer first
4. 28foundation · easy

   In Azure Synapse Analytics dedicated SQL pools, what is the difference between table partitioning and table distribution?

   • ADistribution splits data across physical nodes; partitioning splits data within each distribution
   • BPartitioning is used for rowstore tables; distribution is used for columnstore tables
   • CPartitioning and distribution are interchangeable terms
   • DPartitioning splits data across physical nodes; distribution splits data within a table
   Reveal Solution Discussion Select an answer first
5. 29foundation · easy

   Which scenario is a common reason to implement table partitioning in Azure Synapse Analytics?

   • ATo eliminate the need for distribution keys
   • BTo increase the number of distributions in the table
   • CTo automatically compress the table data
   • DTo enable fast deletion of old data by dropping entire partitions
   Reveal Solution Discussion Select an answer first
6. 30application · medium

   A healthcare analytics company stores patient encounter records in Azure Data Lake Storage Gen2. They use Azure Synapse Serverless SQL to run analytical queries that filter by encounter_date and by facility_id. The current layout is /encounters/2024/05/15/data.parquet. Queries are slow. You need to optimize the layout for analytical workloads. What should you do?

   • AKeep the current layout and rely on Synapse Serverless SQL's automatic optimization.
   • BChange the layout to /encounters/facility_id=123/date=2024-05-15/data.parquet.
   • CConvert the files to JSON format for better query performance.
   • DCreate an external table with a partition on encounter_date only.
   Reveal Solution Discussion Select an answer first