3 Replies Latest reply on Oct 9, 2013 7:51 AM by Matt Lutton

    packaged workbook on tableau server

    Robert Sutter

      Is there a difference in performance between a packaged workbook running in Tableau Server vs. A workbook with a data source published separately?

        • 1. Re: packaged workbook on tableau server
          Matt Lutton

          There can be;  it depends on the data source and whether you are using a live connection or an extract.  If I publish a data source to Tableau Server as a published data source, and I connect a workbook live to it, it will be much slower than if I create a local copy of the extract and publish the workbook with the extract.

           

          I'm not sure if that helps or not, but a lot depends on the type of data source, the amount of data in that source, and whether you are connecting live or using an extract.  Where the data source is stored doesn't seem to make much of a difference, its the type of connection used that seems to matter most.

          • 2. Re: packaged workbook on tableau server
            Robert Sutter

            Yes I should have clarified.

             

            The twbx I'm referencing is based on an extract. So I guess what I'm asking

            is will the workbook load faster in Tableau Server if the extract is

            separate vs. a packaged workbook.

             

            Also any links on optimizing visualizations would be helpful. It'***** and

            miss finding them on this forum.

            I know I've run across a few where Lookup formulas are used to speed

            calculations but I can't seem to find them.

             

             

            On Wed, Oct 9, 2013 at 10:09 AM, Matthew Lutton <

            • 3. Re: packaged workbook on tableau server
              Matt Lutton

              This is a good place to start:

              Performance | Tableau Wiki | Drawing with Numbers

               

              If you publish the Extract and utilize the published data source instead of a local extract, you will be using a "live connection" to the extracted data, which could be much slower--the only way around that is to create a local copy of that extract, which is essentially the method you're already using.  I would stick with your current method.

               

              If there were a way to publish an extract, and allow multiple workbooks to connect to that extract while retaining the performance benefits of extracts, that would be a huge improvement.  For now, publishing Data Sources is more of a way to do data modeling across organizations, so many users can connect to pre-set data sources with calculations, aliases, etc. already created for them.

               

              Hopefully what I am saying is making sense.