Understand the Optimization Techniques Used by Databricks to Ensure the Performance of Queries on Large Datasets (data Skipping, File Pruning, Etc.). DATA-ENGINEER-PROFESSIONAL Practice Questions (Page 5)

Questions 21–25

1. 21foundation · easy

   Which Databricks feature automatically reduces the number of small files during writes?

   • ADelta Live Tables
   • BPhoton
   • COptimized writes
   • DAuto Loader
   Reveal Solution Discussion Select an answer first
2. 22expert · hard

   A Delta table has been Z-ordered on a frequently filtered column, and statistics are up to date. However, a query filtering on that column still scans all files. The table has a high number of small files. What is the most likely cause?

   • AStatistics are not collected for Z-ordered columns
   • BThe small file count prevents effective data skipping because each file's statistics are not selective enough
   • CZ-ordering only works on partitioned tables
   • DThe query uses a range predicate instead of equality
   Reveal Solution Discussion Select an answer first
4. 23foundation · easy

   How does Databricks maintain statistics for data skipping on Delta tables?

   • AStatistics are manually collected by running the ANALYZE TABLE command after every write
   • BStatistics are stored in a separate metadata store outside the transaction log
   • CStatistics are derived from the file names and directory structure
   • DStatistics are automatically collected and stored in the transaction log when files are written
   Reveal Solution Discussion Select an answer first
5. 24expert · hard

   A Delta table has been optimized with Z-ordering and Bloom filter indexes. However, a query filtering on a column that is neither Z-ordered nor indexed still scans all files. The table has up-to-date statistics. What is the most likely reason data skipping is not working for this query?

   • AStatistics are only collected for Z-ordered columns
   • BBloom filter indexes are required for data skipping
   • CThe table is too large for data skipping
   • DThe column's values are distributed such that min/max statistics do not exclude any file
   Reveal Solution Discussion Select an answer first
6. 25expert · hard

   A Delta table is written by a streaming job that appends small batches every minute. The table has Auto Optimize enabled, but query performance is still poor because data skipping is ineffective. What is the most likely reason?

   • AAuto Optimize compacts files but does not cluster data by the filter columns
   • BThe table is partitioned incorrectly
   • CAuto Optimize does not work with streaming writes
   • DStatistics are not collected automatically
   Reveal Solution Discussion Select an answer first