A daily revenue dashboard reads from a lake fact partitioned by user_id, producing 10 million tiny p
A medium Pipeline Design interview practice problem on DataDriven. Write and execute real pipeline design code with instant grading.
- Domain
- Pipeline Design
- Difficulty
- medium
Problem
A daily revenue dashboard reads from a lake fact partitioned by user_id, producing 10 million tiny per-user directories. Daily-filter queries cannot prune because the partition key does not match the filter. Apply the section's partitioning framing and replace the wrong partition key with a section-named canonical date key. The high-cardinality key belongs on a clustering key, not a partition key.
Practice This Problem
Solve this Pipeline Design problem with real code execution. DataDriven runs your solution and grades it automatically.