Identify Poorly Performing Queries in the Databricks Intelligence Platform, Such as Query Insights, Query Profiler Log, Etc. DATA-ANALYST-ASSOCIATE Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   What is 'data skew' in the context of query performance?

   • AA situation where data is evenly distributed across all tasks
   • BA situation where data is encrypted
   • CA situation where data is stored in multiple formats
   • DA situation where some tasks process significantly more data than others
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   When using Query Insights to find poorly performing queries, which filter would help you focus on queries run by a specific analyst?

   • AFilter by data source
   • BFilter by query ID
   • CFilter by user
   • DFilter by warehouse size
   Reveal Solution Discussion Select an answer first
4. 8foundation · easy

   How can Query Insights and the Query Profiler log be used together to diagnose a performance bottleneck?

   • AQuery Insights automatically fixes the query, and the Query Profiler log verifies the fix
   • BQuery Insights provides the SQL text, and the Query Profiler log provides the results
   • CQuery Insights stores the query history, and the Query Profiler log stores the data
   • DQuery Insights identifies the query, and the Query Profiler log provides detailed execution statistics to pinpoint the bottleneck
   Reveal Solution Discussion Select an answer first
5. 9expert · hard

   A data analyst is examining the Query Profiler log for a query that aggregates data by a 'country' column. The log shows that one task in a stage takes 10 minutes, while all other tasks take less than 1 minute. The stage's overall duration is 10 minutes. What is the most likely cause, and what is the best first step to address it?

   • AA configuration issue, and the analyst should increase the number of shuffle partitions.
   • BA CPU bottleneck on one node, and the analyst should check the node's resource usage.
   • CA network bottleneck, and the analyst should check the cluster's network throughput.
   • DData skew, and the analyst should investigate the distribution of the 'country' column to see if one country has a disproportionately large number of records.
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   A data analyst is looking at the Query Profiler log for a query and wants to see the total amount of data read from disk and the total amount of data written to disk during the query's execution. Where in the Query Profiler log should the analyst look?

   • AThe shuffle map, which shows the data moved between stages.
   • BThe task-level metrics, which show the duration of each individual task.
   • CThe stage-level metrics, which include input and output bytes for each stage.
   • DThe execution plan, which shows the sequence of operators.
   Reveal Solution Discussion Select an answer first