Best Practices - Efficiently Handling Multiple Data Sources

Goal

When creating workbooks that have multiple data sources, many complex joins and calculations, learn best practices to make them run smoothly.

Learn

Use these recommendations to handle multiple data sources:

  • Use an import job instead of a data link for better performance
  • Set up time-based partitioning and select smaller sets of partitions whenever possible 
  • Append instead of replacing in imports 
  • Separate data layers from analysis across different workbooks (one-time joins, unions, cleansing, etc.)
  • Separate joins into their own workbooks to avoid re-running them when actual analysis changes

Further Readings

Workbook Optimization Guide
Artifact Naming and Structure Best Practices