Implement a Partition Strategy DP-203 Practice Questions (Page 5)

Questions 21–25

1. 21foundation · easy

   For an analytical workload that frequently aggregates sales by region and month, which partition key would most improve query performance?

   • APartition by sales amount range
   • BPartition by region
   • CPartition by month
   • DPartition by region and month
   Reveal Solution Discussion Select an answer first
2. 22application · medium

   A retail analytics team stores product sales data in Azure Data Lake Storage Gen2. The data is used for both daily reporting and ad-hoc queries that filter by product_category and by sale_date. The current layout is /sales/2024/05/15/sales.parquet. Queries are slow. You need to identify whether partitioning is necessary and implement it. What should you do?

   • APartition only by sale_date, ignoring product_category.
   • BConvert the files to Avro format for better performance.
   • CPartition by product_category and sale_date, e.g., /sales/category=electronics/date=2024-05-15/sales.parquet.
   • DKeep the current layout because partitioning is only needed for very large datasets.
   Reveal Solution Discussion Select an answer first
4. 23application · medium

   A media company stores raw clickstream logs in Azure Data Lake Storage Gen2. The logs are appended continuously and total about 2 TB per day. Data scientists run exploratory queries that often filter by date and by user segment. The current layout is /logs/clickstream_<timestamp>.json. Queries are slow and expensive. You need to improve query performance and reduce cost. What should you do?

   • AConvert the JSON files to Parquet and keep the current folder structure.
   • BPartition the data by date and user segment in the folder structure.
   • CUse Azure Data Factory to copy the logs into a single large file per day.
   • DEnable hierarchical namespace on the storage account.
   Reveal Solution Discussion Select an answer first
5. 24expert · hard

   A healthcare company stores patient records in Azure Data Lake Storage Gen2. The data is partitioned by year and then by month. Each day, about 5,000 small JSON files are written. Data scientists run queries that filter on a specific patient ID and a date range of one year. Queries are slow because they scan many small files. You need to improve query performance while minimizing changes to the ingestion pipeline. What should you do?

   • AUse Azure Data Factory to compact small files into larger Parquet files after ingestion.
   • BEnable Azure Data Lake Storage Gen2 hierarchical namespace.
   • CAdd a patient ID level to the folder structure: /patients/{patientId}/{year}/{month}/
   • DChange the partition structure to month first, then year.
   Reveal Solution Discussion Select an answer first
6. 25foundation · easy

   In a streaming data pipeline, what is the main purpose of partitioning the event stream by a key such as device ID?

   • ATo reduce the total number of events ingested
   • BTo guarantee that all events are stored in a single file
   • CTo eliminate the need for checkpointing
   • DTo ensure events from the same device are processed in order
   Reveal Solution Discussion Select an answer first