Azure SQL Database: Columnstore indexes

with Lara Rubbelke, Davide Mauri

Davide Mauri is back with Lara Rubbelke to discuss columnstore indexes. In a typical relational database data is stored per row. This means that all row columns are saved together, which is great in all those cases where you want to retrieve all the information related to a specific row. But what if you want to have all the data stored in a subset of columns but for all the rows? That would be complex and slow—unless data was saved per-column instead of per-row. That's exactly what columnstore indexes in Azure SQL Database enables you to do.


  • 00:00 - Introduction
  • 01:00 - Columnstore overview
  • 07:00 - Demo: index creation
  • 10:14 - Demo: querying with columnstore
  • 13:53 - Demo: batch mode with LAG operator
  • 15:00 - Wrap-up


Azure SQL Database