Tableau Queries Better With Short & Wide or Tall & Thin Tables in SQL Server?

Question

I'm trying to improve the query time for Tableau when it queries data from the table in my SQL Server database. When considering # of rows vs. # of columns, should I have a short & wide table or tall & thin?

This resource has done some research proving that the tall & thin approach is better: http://www.powerpivotpro.com/2011/08/less-columns-more-rows-more-speed/, but I wanted a few more sources of info before I started the formatting of the tables.

Would multiple tall & thin tables that are joined be less performant than 1 short & wide table when Tableau is pulling data from it to make visualizations?

Answer 1

• Tableau can go better with tall data than wide data. This is because it does better when all the numbers to graphs are in the same column. Tableau performs well with a category field. The category can do a variety of different things when used in the boxes for "color", "detail" (breakdown by), "label".

• Tall example

• date, category, value

• date1, sales, 10

• date1, revenue, 5

• date2, sales, 12

• date2, revenue, 6

• Wide example

• date, sales, revenue

• date1, 10, 5

• date2, 12, 6

• Refer to below links for Excel data reshaper

http://www.tableausoftware.com/about/blog/2012/03/reshaping-data-made-easy-16353

http://kb.tableausoftware.com/articles/knowledgebase/addin-reshaping-data-excel

• Tall and thin is usually a good way to go with Tableau to simplify the logic, but there are exceptions depending on your data and your goals. If your aim is to improve performance, nothing beats running experiments. Try the Tableau performance recorder to see the query and the response time. Other things to investigate for performance tuning include data extracts and/or tuning your indices

That’s all for now. But, if you would like to learn more about Tableau, check this Tableau Desktop Specialist Certification course by Intellipaat.