4 Replies Latest reply on Apr 25, 2018 6:13 AM by Jatinder Singh

    Performance Issues: Massive Dataset 100M+

    Jatinder Singh

      Hi guys,

      I am with an amazing challenge of preparing a tableau dashboard to cosume about +100M rows from a Single Table of SQL Server Database.

       

      As the data volume is pretty High I am facing the performance issue. So far I have performed the following optimization:

      - context filters -> to fetch a smaller data set before initiating the calcs with the  indexing field.

      - reduced and converted string fields to Numeric and Boolean where ever possible.

      - A bridging table which contains all the distinct possible values to fetch data from the bigger data set.

      - Data extracts instead of the live connection

       

      All these optimization have increased the performance significantly, but still not acceptable performance as every time an extract based query is launched it takes about 20-30 seconds.

       

      Any further suggestions guys  ?

       

      Next step I am thinking is to break the single table into smaller entities and create joins in tableau, but i am not sure if this would boost the performance.