Hey guys,
Basically, I've asked my question in the title.
I had an issue the other day. One of my colleagues approached me because one of our reporting servers had gone down and they tracked the issue to the Tableau server.
I realised straight away why this issue occurred. In hindsight, a full refresh of a large dataset every 15 minutes isn't a great idea. I now intend to try out using incremental refreshes and reducing it to once a day.
Will this really drastically reduce the load on the database server though. Obviously reducing it to once a day will help, but I don't entirely understand the mechanism behind the incremental refresh.
Is it still querying the entire database but only extracting a portion? Thinking about it, is it the smaller extract which will create a lesser load?
Hi Steven,
The incremental refresh uses a reference column to select the data so it won't select all the data from the source.
Tableau Data Extracts - Tips, Tricks and Best Practices | Tableau Software
You can also inspect the logs for the data source to see the queries made by the Extract refresh that could then be used to add indexes to further optimize performance.
Hope this helps-
=Virginia
