Identify Poorly Performing Queries in the Databricks Intelligence Platform, Such as Query Insights, Query Profiler Log, Etc. DATA-ANALYST-ASSOCIATE Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   While reading the Query Profiler log for a slow query, an analyst notices that one particular stage has a very high duration compared to all other stages. Within that stage, a few tasks have significantly longer durations than the rest. What is the most likely performance issue indicated by this pattern?

   • AA bottleneck in the driver node, slowing down the entire query.
   • BA data skew problem, where some partitions have much more data than others, causing a few tasks to take much longer.
   • CA configuration issue where the cluster is too small, causing all tasks to run slowly.
   • DAn excessive number of shuffles, causing network overhead across all tasks.
   Reveal Solution Discussion Select an answer first
2. 12application · medium

   A marketing analyst is using Query Insights to troubleshoot a slow dashboard. The analyst suspects the issue is caused by queries run by a specific user on a specific date. How can the analyst narrow down the list of queries in Query Insights to find the problematic ones?

   • AOpen the Query Profiler log for the dashboard and look for the user's name in the log entries.
   • BNavigate to the cluster's event log and filter by the user's name and date.
   • CUse the filter options to select the specific user and date range, then sort by execution time.
   • DExport all query data to a CSV and use a spreadsheet to filter by user and date.
   Reveal Solution Discussion Select an answer first
4. 13application · medium

   A data analyst is comparing two queries in Query Insights. Query A has a total execution time of 5 minutes, and Query B has a total execution time of 2 minutes. However, Query A's 'Execution Time' is only 1 minute, while Query B's 'Execution Time' is 1.5 minutes. What can the analyst conclude from this comparison?

   • AQuery A is spending most of its time waiting for resources, while Query B is spending more time actively computing.
   • BQuery B is performing more work because it has a higher execution time.
   • CQuery A is more efficient because it has a lower execution time.
   • DQuery A has a data skew problem, while Query B does not.
   Reveal Solution Discussion Select an answer first
5. 14expert · hard

   A data analyst is investigating a query that joins a large fact table with a small dimension table. The Query Profiler log shows that the join operation is causing a massive shuffle, and the stage duration is very high. The analyst wants to reduce the shuffle and improve performance. Which approach is most likely to be effective?

   • ARepartition the fact table by the join key before the join.
   • BIncrease the number of shuffle partitions to distribute the data more evenly.
   • CUse a broadcast join hint to send the small dimension table to all executors, avoiding the shuffle.
   • DUse a larger cluster to provide more resources for the shuffle.
   Reveal Solution Discussion Select an answer first
6. 15foundation · easy

   In Query Insights, which metric indicates the total time a query spent waiting for resources before execution?

   • AResource usage
   • BWait time
   • CRows scanned
   • DExecution time
   Reveal Solution Discussion Select an answer first