Home / Data Engineer Associate / Objective 6.3

Domain 6: Troubleshooting, Monitoring, and Optimization · 10% of the exam · Full exam outline

6.3 Identify common performance bottlenecks such as data skew, shuffling, and disk spilling by interpreting stage-level metrics in the Spark UI.

Practice questions

Question 1medium

In the Spark UI, the Summary Metrics for a long-running stage with hundreds of tasks show a 75th percentile task duration of 40 seconds and a Max of 9 minutes. What does this pattern most likely indicate?

Included in Master: master every domain.See plans
Question 2medium

The slowest stage of a batch job shows spill statistics in its stage details, while the task durations in its Summary Metrics are evenly distributed. Which change should the engineer try first?

Included in Master: master every domain.See plans