Hey Kevin,

It definitely does if the data is not extracted. In fact, the database setup itself can affect the performance if the data is not extracted - for example, a slow view in SQL server will perform worse than a fast view in SQL Server.

I believe Cubes, in particular, cannot be extracted (someone who uses cubes correct me if I'm wrong), so could be much slower than a static excel file.

I would recommend always extracting the data if possible.

Hope this helps!

- Derrick

one of the example of data source related

Hortonworks and Tableau is slow

What matters is:

1. Data locality: local files or databases that are on you machine will be faster than something on your network or in the cloud
2. Data size: obviously, less data we need to query or bring the better.
3. In general Tableau doesn't move or process lot of data and this task is offloaded to database engines and yes, some databases are faster than the others

To answer your specific question: given same amount of data - cube should be faster however I find this comparison quite incorrect. Like comparing truck with carton box