Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
01-14-2025 11:34 AM
Well, in the end, it was caused by skewed data. Document_ID was -1 for returns in sales, so a big part of the table was filled with -1 values. Adding an extra column to the merger solved the problem.
This article helped me a lot:
https://www.databricks.com/discover/pages/optimize-data-workloads-guide#data-skewness
